Thursday, December 16, 2010

Transaction Analysis

This is an a note to RSS readers that a new transaction analysis has been published at the main BaseballNotebook.com website.  It is not viewable in this format.

Friday, December 10, 2010

Transaction Analysis

A new transaction analysis report has been published at the main BaseballNotebook.com page.  It is not visible in this format.

Thursday, December 09, 2010

New Transaction Analysis

Please note that two new transaction analysis reports have been published at the main BaseballNotebook.com page.  They are not viewable through this feed.

Tuesday, December 07, 2010

Adrian Gonzalez, Shaun Marcum, Mark Reynolds

Baseball Notebook has resumed publishing for the 2011 season and pre-season! Today's blog entry, which contains charts that do not display properly here, can be found through this link or by visiting the main BaseballNotebook.com page.

Thursday, June 10, 2010

The Significance of Strasburg's Debut

There are a couple of easy mistakes to make in forecasting that I often warn about:

- Never be fooled by a small sample.
- A great outcome that gets our attention is not a randomly selected sample.

I'm about to make a couple of exceptions for good reason and I will explain why.  A single game performance can be so outstanding that it is extremely unlikely that it can be accomplished by the vast majority of players.  Also, even a seemingly small sample can be large enough to tell us something in the rarest of cases.

Let me give you an example: Suppose we are standing at a golf course waiting our turn to golf on the first hole.  We are observing a player ahead of us whom we know absolutely nothing about other than what we can visually observe.  He reaches into his bag, pulls out his driver on a hole that's over 300 yards away, takes a practice swing and then confidently swings and we watch as the ball magically soars, lands on the green and rolls right into the hole for a hole in one.

Believe it or not, that sample size of just one means something and tells us quite a bit.  We can't be positive but the chances are extremely high that the golfer we've just seen did not just take the first swing of his life.  It's possible but unlikely.  He may not be a great or a pro or even a top amateur but we've at least narrowed the likely range of skill that he has by observing a single swing.

That the hole in one got our attention is dangerous... We noticed the player because he did something extraordinary the first time we saw him.  Remember Karl "Tuffy" Rhodes hitting 3 home runs on opening day a number of years back?  Better yet, how about Mark Whiten's 4 home run game?  They certainly got our attention and if we were to examine the odds of players doing what these players did, they would be low and we could easily make false conclusions about their ability.

But there is a point where someone does something that is so unlikely that it is more likely that they have unusual ability than unusual luck.  That brings me to Stephen Strasburg's major league debut.

Not only did the debut live up to the hype but it exceeded it.  Strasburg didn't just look fantastic and from a non-statistical perspective, his stuff was electric.  He struck out 14 batters of the 24 he faced but in doing so, walked a grand total of none.  That's extraordinary.  Using tools available at baseball-reference.com (a site I highly recommend and which should be a frequent stop on every reader's Internet browsing), I found a grand total of 20 games since 1968 where a pitcher struck out at least 14 batters without walking a batter.  The first few names since 1968 were Sam McDowell, Luis Tiant, Vida Blue and Frank Tanana, all of whom had had excellent careers.  In the 1980s, the names were Dwight Gooden (twice), Mark Langston and Sid Fernandez, again all pitchers who had strong careers.  Since 1990, the names are those you would expect to see: Pedro Martinez (twice), Randy Johnson (twice), Roger Clemens, Curt Schilling, John Smoltz, Johan Santana and Mike Mussina make the list.  Then there are a couple who may not have had the careers of these others but still managed to accomplish the feat in Eric Bedard and Mark Prior.  That's it since 1968.

But again, we run into the problem of singling out something that got our attention rather than a random sample of a game.  So, let's consider this from a different angle.  What if we could simulate a large number of random games made up of exactly 24 batters faced?  We could try different theoretical levels of strikeout ability and see how often our imaginary pitcher strikes out at least 14 batters.

Let's start with the average National League pitcher from 2009, who would strike out about 18% of the batters he faced.  With our ability to run high speed simulations of blocks of 24 batters faced, I ran 10 million of these and at the end of 10 million "games" the average NL pitcher from 2009 struck out at least 14 out of 24 batters in 149 of the 10 million games or about 1 in every 67,114 games.

In other words, it's possible that Stephen Strasburg is not yet at least an average National League pitcher but based on a single start, an incredibly small sample, we can already say that it's extremely likely that at least where strikeouts are concerned, that one game demonstrates that he's already above average by NL standards.

But you probably already thought that.  Let's raise the bar.  Let's give him the theoretical ability of a pitcher who strikes out about one out every 4.5 batters he faces or about the rate Josh Beckett had in 2009.  When we ran ten million games of that sort of pitcher, the 14 strikeout threshold was met or exceeded 1,491 times or about once every 6,706 games.

Let's go even further: Let's give our imaginary simulation pitcher the ability to strike out about one out of every 3.5 batters he faces or about what Tim Lincecum did in 2009.  When we ran that simulation, again ten million times, our imaginary pitcher of Tim Lincecum's approximate strikeout skill managed to achieve the 14+ strikeouts in 24 batters faced mark 21,881 times or about one out of every 457 times.

I still hesitate if not outright reject making conclusions based on a performance that gets our attention simply because of its excellence but this isn't just one night where a pitcher struck out 14 batters and walked none.  This was the top-ranked pitching prospect in the world making his first major league appearance.  That's not necessarily a random sample and we could have easily resolved to do this sort of analysis on whatever he happened to achieve in his first game.

My suspicion is that Strasburg actually has nearly proved something beyond an acceptable degree of confidence and that is that he is an even better strikeout pitcher already than I expected he would be at this early stage and my next projection revision will reflect this.  His minor league numbers this year didn't even hint at this sort of performance and I'm expecting that both he and his team were (and maybe still will be) controlling his effort enough to not over-exert himself.

One other notion that I think deserves mentioning is that we hear some reminding us that facing the Pittsburgh Pirates did not offer Strasburg an average major league opposition.  While that may be true, the Pirates may be a low-average team (worst in the majors at .238) but they are actually not that bad at making contact, at about a 79% rate, almost exactly the current major league average, which is about 79.6% so far this season.  So, Strasburg's strikeouts achievment is not significantly lessened by the Pirates having been his opposition in his first game.

Anyway, despite everything I often write about small samples and overexcitement about huge single game performances, I believe that Strasburg's first game is that rarest of exceptions.   I don't think it proves that he's going to strike out double digits in every game but it does demonstrate a significantly superior strikeout potential than I had previously expected so early in his development.  If you didn't watch his first game, I highly recommend you tune in to watch his second start.  It's not about whether he can match what he did in game one as much as you really need to see his stuff if you enjoy the game.

Wednesday, June 02, 2010

Trading and the Gambler's Fallacy

We're hitting that point in the year where the contenders and the also-rans in fantasy leagues are becoming clearly separated and teams are making moves to strengthen their run for a championship or, if they think they can't contend this year, building for the long run.

Occasionally, I pop into message forums to see what the chatter is and of course, I often hear from readers asking about whether they should trade for player X despite a lousy start to the season, and at this time of year, I've personally responded by email to a greater percentage of these emails than usual.  One of the trends I've noticed, and it's not the first time I've seen this, is that there remains a trend for some fantasy leaguers to confuse the law of large numbers with the oft-used phrase "the law of averages" and/or committing the gambler's fallacy.

The law of large numbers does not tell us that a player will do more at some later point to offset the results from the period when he did less earlier.  In that same line of thinking, it does not mean that a player who has started any season above his real ability will now necessarily do worse later so that his final totals somehow arrive at his real ability.  The larger the sample size, the more an entity tends to perform closer to its real ability but there is not an offset effect.  In other words, let's return to our favorite coin flipping example.  Assuming for this example a fair coin, let's say the coin goes 7 for 10 in April on tails.  This does not mean that the coin should now go 3 tails for 10 flips in May to "balance out" things.  The coin still has a 50% chance of coming up heads or tails.  If we flipped the coin another billion times or so, that 7 for 10 start will be a long-lost memory because in the grand scheme of tossing the coin a billion times, the sample size of 10 would have a virtually non-existent effect on the final totals.

The gambler's fallacy, as it is often called (and worth Googling if you've never read about it), is the belief that somehow things will balance out as if the deviation from real ability so far that we've witnessed will be offset by a reverse deviation in the opposite direction.  I remember years ago reading an example that went something like this: If Wade Boggs is hitting .200 at the end of April, then you should trade for him because you know his final totals will be around .330-.360 at the end of the season and you'll end up with the portion of the season that helps him achieve these totals.  Wrong.  You should have traded for Wade Boggs because his ability was so good at that time that you'd end up with a lot of value but if he were to hit .200 through the first month and his real ability was .360, it did not mean he was more likely to hit .380 the rest of the way to make up for the slow start.  The most likely outcome would still be .360.  Now, if that .200 start convinced you he was declinined and you decided he was actually becoming a .350 hitter instead of a .360 hitter, well then you would project him to hit .350 the rest of the way, presuming neutral other variables that influence the remaining outcome.  You see this sort of thinking in so many circles about expecting reverse effects to balance things out.  The roulette player who has witnessed five blacks in a row believes that red is now due to come up and when red finally, and inevitably eventually comes up, he exclaims "I told you!" when you can keep doing this forever until you hit red.

I wanted to write about this today not only to deter trading for players because you believe they're about to perform beyond their ability to balance out a slow start but also to deter you from dumping players who have started out hot just because you think the slump is coming that will offset it.  By all means, winning at fantasy baseball absolutely requires that you exploit the incorrect perceptions generated by a player's season-to-date.  If you think a pitcher's real hidden ERA skill is around 5.50 and he has an ERA of 3.00 so far on the season, you have to make your move and maximize your return.  But if you believe his real hidden ERA ability is 3.75 for that same pitcher, it doesn't mean that his hot start should have you believing he's suddenly going to collapse performance-wise just because he's already experienced the good part of his season.

You have to assess the player's remaining expected value but always avoid incorrectly applying the season-to-date numbers to make "reverse calculations" that try to say "okay, because he's already got an ERA of 6.00, that means his ERA will be 2.50 the rest of the way if he's to land at his real 4.00 ability by the end of the season."  It doesn't work that way.

Take the often-discussed Javier Vazquez for example.  He's much younger than you might believe if you read speculation about him being at the end of the line (Vazquez turns thirty-four later this month) and while we've been forced to adjust our forecasts to reflect that his terrible start jeopardizes his spot in the rotation for a while, it also doesn't mean that we expect him to suddenly go on some 1988 Orel Hershisher-like stretch to offset that start just because we believe his ability is better than he's shown.  His season is going to look bad for a long time and there's no erasing what's already in the books.  So, when/if you trade for him, and he's an example of a player who has become surprisingly easy to acquire relative to just a few months ago, don't be banking on him pitching above his projected ability level the rest of the way.  Just trade for him because you share our belief that he's a better pitcher than we've seen and his projected remaining value can be of use to you.
Now, this is where there is another exploitation opportunity in trading right about at this time of year.  That is, there are people who do make the gambler's fallacy and they're playing against you in your fantasy league.  There are fantasy GMs, potential trading partners, who actually expect players to perform outside of the ability that even they project because they commit the gambler's fallacy.  These are the guys who will actually be anxious to trade a Jose Bautista because they think that his unusually strong start is a fluke and that Bautista will come crashing down later in the year.  Well, even if we agree that Bautista is overachieving, that doesn't completely negate his value and we will still be forecasting him to perform according to our revised estimate of his long run ability, not less than that simply because he had a hot start.  Maybe there are fantasty trading partners who don't dump their players like this but who are too smart for their own good and want to trade for players whose decline to date represents a real problem but for whom hope is held out that the rest of the season will be a bounce-back beyond their projected ability to balance things out.

In short, always remember that when you're making a trade, you need to be assessing what you believe the player's real ability is, both for the short-term and long-term.  Certainly, no one can deny that what a player has done so far this season informs our opinion of that but it's crucial to use that season-to-date performance properly, to influence our evaluation of ability and not to make some sort of offset estimate that tries to say "he's already hit 10 of his projected 15 home runs so a big drop-off is coming" or variations along those lines.  Doing that is not only a quick path to frustration but it also will have you waiting for streaks and slumps that even if and when they come, will be fluke more than fact and will further confuse making accurate evaluations of a player's real hidden and long-term sustainable skill level.

Wednesday, May 26, 2010

Good Luck, Bad Luck

We've been working recently on introducing a report similar to one we used to publish a few years ago called "Good Luck, Bad Luck" which, if all goes as planned, will be part of a larger effort on our part to consolidate much of the most useful statistical info into a new-style newsletter, enabling readers to get items such as the weekly depth charts, week ahead reports and others possibly as early as Friday rather than having to wait for the weekends, when they're available at the site for all to see.  We're not sure of the details yet but at least elements of this old luck report will likely be a part of this effort and we're aiming for something that gives us a bit more flexibility on the presentation, such as an Adobe .pdf style newsletter.  Details will follow, probably at some point in June, but we think this could greatly enhance the free side of our site (the subscription side remaining the weekly-revised statistical player projections and ranking sheets).

As we've been toying with some of the information to include in such a report, we recently produced one of those old-style luck reports on the current year stats, through play completed last Thursday.  It revealed a few interesting things that reminded me of why I really liked having that information easily available.  Here are a few items that jumped out, again through play completed last Thursday:

Wes Helms has been picking up singles on 36% of the balls he has put in play, through play completed last week.  To put that in perspective, that's well above the typical league-leading level for any player and Ichiro led all of baseball last year with a 31.5% rate.  In other words, Helms' average so far this season is at least partially the result of some pretty favorable luck.

Put Sterlin Castro into that category too as heading into last weekend, he had been singling on 35% of the balls he had put in play.

Other players who were topping Ichiro's rate of singles per ball in play of 2009 include Elvis Andrus, Jamey Carroll, Mike Aviles, Ryan Theriot and Edgar Renteria.

On the flipside of the equation, there were some players who were having some pretty miserable luck on singles falling in.  Included among them was Aaron Hill, who had an incredibly low 11% of balls in play falling in for singles as of the end of last week.  To put that in perspective, a 13% or 14% would normally be the lowest we would see here among the worst player in the league (Carlos Pena was the lowest among qualifiers last year at 13.3%) and Hill himself had a rate of 20.1% last year in this column.  To say Hill has been unlucky would an understatement as there's no way this rate will stay at 11% in the long run even if he has become a worse hitter than his track record implies.

Hill's not alone at achieving incredibly low, and likely unlucky, singles rates so far this year as Carlos Quentin was on the list at 12% and the surprisingly-powerful Jose Bautista is also there at 12%.

On the pitching side of bad luck, Carlos Zambrano has seen 40% of balls in play, other than home runs, fall in for base hits and that rate is so high that there's no way it can't come down if he keeps pitching.  Others on that list included Doug Davis (39%), Bud Norris (38%), Justin Masterson (38%), Brandon Morrow (37%) and Gavin Floyd (37%).

On the lucky side of balls falling in for hits, Livan Hernandez has been not only incredibly lucky but almost historically lucky here in that just 18% of balls in play, other than home runs, have fallen for hits.  This is a rate that is virtually guaranteed to go way up and when it does, the rest of his pitching numbers will fall in line with exactly the type of pitcher everyone already knows he is.  Other names on the lucky list include Jason Vargas (21%), Jamie Moyer (22%), Doug Fister (22%) and Ubaldo Jimenez (22%). One name on the list was a surprise because even showing up here, he's still not having a good season and that is Todd Wellemeyer.  He has seen only 20% of his non-HR balls in play fall for hits, this even as his ERA keeps flirting with the mid-5's.

Anyway, I just wanted to share a few highlights from this report as we're working on something new that will likely include elements of this report.

Wednesday, May 19, 2010

Gomes, Cust, Cleveland infield, Sizemore/Scherzer/Boesch

As I contemplate the next set of major changes for the upcoming projection set this weekend, I wanted to take a moment to highlight a few players who I can already see are poised for some fairly significant upgrades and downgrades in the next set.

One that really jumps out at me is Jonny Gomes.  It's looking like he's established as a regular player in the lineup and while I'm not on the verge of upping his projected batting average much, he has some definite fantasy value.  If he continues to play on a regular basis, unlike what we originally and previously were forecasting, he's the kind of player who could still have close to 20 home runs and a handful of steals in him the rest of the way, this if we increase his projected at bats to around 300+ more from this point up, up over 100 from what we had listed previously.

Jack Cust's also a surprising comeback candidate in that he was essentially let go by Oakland in the spring by being designated for assignment, went unclaimed on waivers, and has now bounced back into a position where it looks like he's about to take over from the struggling Eric Chavez as the regular DH.  Assuming that happens, he's still capable of challenging 20 home runs with regular playing time and is probably still sitting on the waiver wire in a lot of fantasy leagues.

Asdrubal Cabrera's injury, which happened after the cutoff for last week's projection set, threatens to keep him out of action until late July or even August and a major downgrade in the next set is obviously coming.  The injury to Cabrera probably helps Luis Valbuena a little.  Valbuena, who had been playing second base but who can also play shortstop, was on the verge of losing his roster spot until this injury to Cabrera happened and while his job is still by no means secure, he gains at least a little bit of security on the roster side of the equation anyway, even if he isn't starting as often as he was earlier in the season.  Jason Donald, who definitely does get a big bump up as a result of Cabrera's injury, doesn't look to be of too much help here, even with a very slight power/speed potential.

One other transaction that had terrible timing in terms of last week's projection set was the completely unexpected demotion of Scott Sizemore and Max Scherzer by the Tigers.  Both were handed tickets to Triple-A after slow starts and neither has an instant path back to the majors.  There's no Detroit pitcher who's going to move up too much in value in the next set (Armando Galarraga projects out as ordinary here even with having it made it back to the majors) but on the hitting side of the equation, Brennan Boesch may have staying power with Sizemore sent down and he's been playing almost every day the past week or so.  If we upgrade Boesch to full-time playing status, he projects out as a low average (don't believe the fast start - he hit only .275 at Double-A last year), 10-15 home run type with only occasional speed, at least for this season.  If you're wondering how Boesch benefits from Sizemore's demotion, it seems the way things are about to play out is that Carlos Guillen, when healthy, will take over at second base and that opens up a lot of playing time through the DH position for Boesch, who can also play in the outfield.  It certainly seems this is the way the Tigers are leaning as of now anyway.

Tuesday, May 11, 2010

Ol. Perez, Haeger, C. Young

As I consider next weekend's updated projection set, there are three pitchers likely to see a huge dive in projected innings for the remainder of the season, all of them National Leaguers.

Oliver Perez is supposedly keeping his spot in the rotation even after his latest disaster, which saw him walk 7 batters in 3.1 innings, but I expect keeping him in the rotation is a scramble to get something in return for the big contract the Mets gave him back before the start of the 2009 season.  After a strong 2007 season and a wild but still effective 2008, Perez then signed a three year deal that is paying him about $36 million total over the life of the contract.  Certainly, the lack of control can't come as a surprise and I now believe he's going to end up in the bullpen within the next couple of months.

Charlie Haeger should be so lucky to end up with such a fate as becoming a reliever.  Haeger was placed on the DL yesterday with plantar fascitis in his foot but I'm thinking that the move just delays the inevitable and that is that Haeger is going to end up losing his roster spot entirely.  In Saturday's outing, just after the cutoff point for our latest projection set, he didn't retire a single batter, walking three and giving up five runs, bringing his ERA up to 8.49 on the season.  If your fantasy league allows you to take players off the roster only when they end up on the DL, now is your chance to cut him.

One name who always seems to come up in the "falling short of expectations" category is San Diego pitcher Chris Young.  We listed Young not only as a very risky player prior to the start of this season but in the numeric version of our risk ratings, he was listed as the 19th riskiest pitcher in all of baseball.  He's living up to that as his return from the DL just keeps getting pushed back and given the latest news the past day or two, I'll be surprised if he's pitching again in the majors before mid-June.  He's poised for a massive downgrade.

In fact, what separates Young from the other two here is that I believe Young can pitch well if he can get active and is probably the only pitcher of the three worth stashing if you can, especially because of the home park.  That still won't stop me from downgrading him to perhaps 70-85 innings projected in the next projection set.

Wednesday, May 05, 2010

The Heavy Weight of April

As we get deeper into the season, this will likely be the final time I get into this topic about hot/cold starts.  I know I'm not the first to say it but it can't be said enough that April absolutely distorts our perception of a player's season.  John Benson used to write about this concept in his one of his old Rotisserie A-Z books and I really like this way of looking at it: At the end of April, what a player did in April is 100% of their seasonal performance.  At the end of May, April is 50% and May is 50%.  At the end of June, April is 33%, May is 33% and June is 33%.  In other words, the player who has a great June but started the season slow gets attention for a hot streak but that's all it will be described as.  If he did it in April, he gets credit for having a great season that gradually cools off to more ordinary levels.  And so on... In other words, April is always there reminding us of how slowly or quickly a player started the year.  A player who hits .400 in April and .200 in May is thought of to be having a good season, even at the end of May, because he "is" hitting .300.  A player who has those exact two same performances but in reverse order, isn't thought of be to having a good season until at least late May, at which time he's viewed as having recovered from a slow start.

I say all these fairly obvious things because I have noticed a theme among those who panic after April performance (and more rarely, celebrate too quickly).  I alluded to this in a couple of previous posts but the short of it is that there are some who want to estimate a player's ability based on a month of play and you simply can't do that.  Granted, a month can definitely tell you something and it's why performance to date still does affect our projected ability to a degree.  As far as the year 2010 is concerned, that one month seems like the only reliable information you have and 2009 seems like ancient history.  The temptation is to favor the sample that is recent over the one that is substantial and finding just the right balance of attention for both is the true battle of every prognosticator.

Just for the fun of it, let me float a couple of possible performances here, each representing a month's worth of performance.  These weren't deliberately selected by digging to find the most extreme examples.  I just picked the first couple I happened to land on:

Player 1: 98 AB, .378 Avg, 2 HR, 19 BB, 26 K, 9 SB
Player 2: 109 AB, .239 Avg, 2 HR, 9 BB, 35 K, 3 SB

Just looking at these two lines above, can we make any real conclusions about either player?  Certainly, we know that both are capable of stealing bases and taking walks but beyond that, is it reasonable to speculate with any degree of certainty based on this small sample that Player 1 is a superior player to Player 2?  How wide would the margin of error be if we attempted to estimate each player's specific skills based on just the information we have?  Well, as it turns out, they're actually the same player from the same season.  Player 1 is David Wright's performance in May of 2009 and Player 2 is his performance in September/October of 2009.  And for what it's worth, this year he's hitting .280 after 116 at bats with 6 home runs, 21 walks, 31 K and 7 stolen bases.

Let's try another, acknowledging now that the theme is that we're actually dealing with the same player each time, selecting two months within the same season.  Recognize this one?

Month 1: 2 W, 3 L, 5.44 ERA, 7 GS, 41.1 IP, 51 H, 23 BB, 37 K
Month 2: 3 W, 1 L, 2.36 ERA, 5 GS, 34.1 IP, 26 H, 10 BB, 33 K

That's Ricky Romero.  Month 1 is actually his final month of the 2009 season.  If we followed the rule of giving more recent data more weight, it should have been more important to us than "Month 2" here which is actually what he did in June.  He's off to a good start this year too but do the above look like the same player?  Of course they don't.  That's because samples of 5-7 games started actually are a small sample and  don't tell the whole story.  Not only is the competition highly variable in smaller samples but the natural course of outcomes says that there will be streaks and slumps to go with good luck and bad luck.  A true .333 hitter, for example, is not going to get a hit on a perfectly even cycle of exactly every three trips to the plate and can look like a .200 hitter in a month.

In some ways, wanting to focus on April data exclusively is akin to wanting to focus on data that only happened after September 1st last year, deciding that what a player has done in the most recent month is of absolute importance to us.  If we did that, we would have concluded at the end of last season that the rest of Ubaldo Jimenez's strong 2009 season was a fluke, this as he slumped to a 4.17 ERA after September 1st.  We would have to believe that Livan Hernandez has suddenly become an elite pitcher based on his start this year.  Honestly, is there anyone out there who believes that Hernandez is a pitcher you should be trading for right now?

The focus of today's piece isn't so much to argue about whether some hot/cold starts represent real changes in ability.  What I want readers to do, and this is crucial to fantasy success, is that you must recognize when a player has turned the corner and is back on normal track.  When that happens, try to not be confused by the April performance being included in the totals.  In other words, it's going to take two months for Javier Vazquez to repair his season no matter how he pitches from now on.  In fact, if Vazquez manages to keep his spot in the rotation (something we're becoming a little skeptical of now as we consider a revision to his games started column) and even if he performs exactly as projected from this point on, after his tenth start of the season, which will come sometime in June, his ERA to date on the season would still be in the high 6's.  To the reader who doesn't remember what we're saying here, if that were the case, it would appear as if he hadn't recovered yet from a slow start when in fact he would have performed from this moment on, in May, exactly as projected.  The season-to-date struggle cannot be removed from the total now and the law of averages does not mean that he will suddenly outperform the projection to balance out a terrible start.

So remember that the earlier a player does something in a season, the more it will skew your perception of his ability, for better or worse.  Keep your eye on (a) his real projected hidden ability, (b) his current role (which is definitely influenced by the hot/cold start) and (c) his health status, which can absolutely influence short-run outcomes, especially in a negative direction.  If you do those things, you'll be continuing to play the long run.

Tuesday, April 27, 2010

Slow Starts that Continue

I mentioned this a couple of weeks ago in this space but I continue to get questions to the mail bag related to players who are off to a cold start.  Some are more strongly-worded than others and we have to be careful not to overreact.  It's a major conclusion to decide that players with long track records are suddenly unable to play the game anymore, especially when we're talking about players in their late twenties or early thirties.  Of course, there's a broader range of ages involved in the slow starts.  Is David Ortiz facing the ultimate decline?  Will Chris Davis ever learn to make contact?

The following seven names seem to be the ones that are showing up the most often in emails sent to our mail bag and I'd like to focus briefly on what each player's start might mean to their remaining season:

David Ortiz: I downgraded Ortiz's forecast in the latest projection set published on the weekend.  No, it wasn't that I was able to make an entirely new conclusion about Ortiz's ability based on fewer than fifty at bats.  It's that I believe the Red Sox are forming an entirely new theory about his ability and what we've seen over the past week is that Ortiz is starting to ride the pine a bit more in favor of other options.  Remember, the DH position is the easiest one to fill because you can use any hitter and the concern with Ortiz is that even with last season's second-half recovery from a terrible start, he did end up with a .238 average by season's end.  I don't believe he's as bad as he's looked this year nor is he as bad as he looked at this time a year ago but it won't matter soon.  I've dropped him to fewer than 300 projected at bats the rest of the way, or about 200 fewer at bats than many regulars in the set.

Chris Davis: There's really not much that needs to be said here except that the Rangers were surprisingly quick to give up on him this year, already sending him back to the minors in favor of Justin Smoak.  As good a prospect as Smoak is, his immediate projected ability is unexciting for a first baseman (our new forecast for Smoak is .259, 11 HR, 48 RBI in 377 at bats from this point on).  So, while we had to downgrade Davis significantly to reflect his demotion, we left in about 100-150 at bats to reflect the expectation that he'll climb back up at some point later this year.  This was a pretty quick trigger, no doubt motivated by Davis's continued propensity for striking out, something that the Rangers shouldn't have been surprised at considering they saw it throughout the year last year.  Strikeouts are part of the Chris Davis power package.

Jay Bruce: We continue to expect good things from Bruce and haven't downgraded him too much, even with his miserable opening to the season.  He's rebounded somewhat the past week but is still only up to .215 on the season through play completed yesterday.  The most important two things you must remember about Bruce are that (a) he just turned twenty-three and (b) he was a .240 hitter in the majors coming into this year in more than 750 at bats.  The reason the second item is important is that because hitters very rarely decline at his age and so his average should be going up as he gains experience and continues to mature physically.  The most difficult aspect of our projection is knowing just how much job security he has and we believe that at least for now, he doesn't have to worry there.

Ian Desmond: Desmond was the highlight of a blog entry I wrote back in the opening week of the season and there's no point in restating what I said there.  In Desmond's case, our projection from before the season still stands pretty much and that is that we're projecting he will lose his job at some point this season.  So, in Desmond's case, the slow start matters because we believe that it represents a real problem for his current ability to hit in the majors, enough so that it challenges his job security.  He did have a 2 for 5 night last night, raising his average from .226 to .241, which is exactly our forecasted level of ability for him for the remainder of the season as well.

Carlos Quentin: There's very little chance Quentin will lose his spot in the lineup unless he gets hurt and his slow start likely represents one of those strange elements of streaks and slumps.  His power performance the previous two seasons promise better things and if you can find someone who believes that Quentin's permanently lost it at twenty-seven, get your trading hat on as quickly as you can.

Javier Vazquez: He's been getting knocked around so far this year and the April misery will form part of the final totals, meaning he'll be in a struggle for a while to get his ERA to respectable territory no matter how he pitches from now on.  Still, he's not quite yet in lose-your-spot-in-the-rotation territory and he's definitely a better pitcher than his 9.00 ERA reveals.  Look at the track record and then look at his age (33) and realize that a sudden decline, while not impossible, is unlikely.

Jason Kubel: Again, we come back to the idea that if a player is still relatively young and has a lengthy track record, then the greatest concern about a slow start should be that it hurts his chances of playing.  Kubel may be facing a little of that now as he did get a day off on Sunday but you have to remember that he hit .300 last year, will turn twenty-eight next month and was a career .278 hitter in the majors heading into this season, in more than 1,600 big league at bats.  In other words, the slow start is almost certainly just a fluke.

In short, if you're trying to assess whether a player's start matters, here are the key factors you have to consider:

1. Is he old enough that this could represent a real, permanent decline?  In other words, a player can decline at thirty-four or thirty-five or, in rare cases, even in his early thirties.  Players generally do not significantly decline in their twenties unless some outside factor happens to permanently change their ability, such as multiple injuries or major surgeries.

2. Is the performance this season over enough of a large sample that we can now give little weight to his previous big league performance?  In short, we're asking here whether a player's .150 average over 50 at bats should have our attention more than his .280 average in the majors over 1,500 at bats.

3. Is the player's performance so far enough to jeopardize his role as a regular contributor?  This one's the most challenging and the most important to get right.  With young players, it's common to see benchings or even demotions as a result of bad starts.  With veterans, they have to be a little worse but you do see it and theoretically, just about any player can end up moved from the rotation to the bullpen or given an extended stay as a pinch-hitter when they're performing poorly.  This is the key question to answer in most cases for if you believe in the player's long run skill, the player still needs to get the playing time to demonstrate his real hidden ability.  In other words, if a pitcher gets moved to the bullpen after 40 miserable innings to start the season, it will take an incredible performance in the bullpen, where he'll get fewer innings, to right the perception ship with his manager.

4. Is a potential replacement candidate good enough to keep this player's job?  For example, we're asking here whether Justin Smoak is good enough to offer enough 2010 ability to the Rangers that they will feel justified leaving Chris Davis in the minors.

Those are the main elements I'm considering when I'm evaluating whether a player's slow start really matters for what we'll forecast the rest of the way.

Wednesday, April 14, 2010

Small Samples That Matter

The other day in this space, I reminded readers of the importance of not putting too much weight into the hot or cold start, especially with only a week in the books.  There is an exception, though, and so today I want to talk about the flipside of that equation.  There are times when you do need to seriously consider how well a player has started and those occasions are when a player's manager believes that the player is better or worse than originally expected.  In other words, if a manager or team's front office becomes convinced that a hot or cold start is representative of a player's current ability (or lack thereof), then it can have a significant impact on the player's projected eventual value.  In particular, this happens the most with rookies and to an arguably greater extent, closers.

With rookies, everything the player is doing early is establishing the perception of how ready the player is for the majors.  A rookie hitting .150 at the end of April is at least somewhat likely to lose his job and one hitting that at the end of May is almost certainly headed back to the minors.  Veterans, especially the highest-paid ones in whom so much is invested, really don't have to worry too much about a slow start.  It can happen when a well-established player starts so poorly that he is given the boot but he's not facing the same challenge of proving himself that a rookie is.  With the rookie, losing a job is virtually guaranteed if the player doesn't offer at least some hint of hope by May or June.

In the case of closers, perception is so important because the manager directly controls the save opportunities.  In other words, potential saves are directly linked to what a manager thinks of a player and of the other pitchers on the team.  For example, Jason Frasor was designated Toronto's closer near the end of spring training but has been shaky in a couple of outings so far, this as Kevin Gregg has performed well enough to already potential challenge him for the role even though so little of the season is in the books.  It has me re-thinking forecasted saves for both Frasor and Gregg in the next edition of the forecasts.  Neftali Feliz has been given at least a temporary bump up to semi-closer status in the past few days, this because he's looked so good and Frank Francisco hasn't.  The list goes on and we're constantly evaluating whether our forecasts need to be revised based on the latest news.

While my entry of the other day was emphasizing that the first week or two of the season reveals only a little about a player's real hidden ability, there is no escaping that managers and front offices do put a lot of weight into early season performance, enough so that we must always consider the hot or cold start in the context of how it impacts a player's potential role and job security.

Therefore, in the same way that you can exploit early perception by getting better than par value for players who have been on hot streaks to start the season or acquiring undervalued ones on cold streaks, you can also take advantage of opportunities that will be earned by a player's hot start or the cold start of another teammate.  It doesn't mean you have to believe that a player is actually as good or bad as he has seemed in early April.  You just have to believe that his manager is putting a lot of weight into it and be looking to acquire players who may be on the verge of new opportunities as a result of unexpected performances early in the season.

Sunday, April 11, 2010

Wells, Perception and the Hot Start

I don't want to repeat what I wrote in an extended essay back in May of 2005 so I'll begin by drawing an old article to the attention of newer readers via this link as it covered a lot of what I would be inclined to write today about the order of performances skewing perception.  In summary, that article was a presentation on just how the earlier stats can create confusion about how a player is actually performing during the current season.

Turning our attention to 2010 or if you're not interested in reading that extended essay, there is no more serious a time to apply the "buy low, sell high" mentality than early in the season.  It's incredible how much a player like Vernon Wells has boosted his trade value with just one strong week.  If he had hit four home runs in a week in, say, June, we'd be talking about a hot streak.  When you hit four home runs in the first week, suddenly the general population is armed only with that performance to analyze your current ability and speculates you're Babe Ruth.

Of course, it wouldn't be fair to pick on Wells for having a strong and well-earned start to the season.  He had wrist surgery during the off-season that may have lifted his power back up somewhat compared to how it appeared last year.  In fact, we are expecting his power to be a bit better this year than last year as we're currently projecting a home run about every 30 times he puts a ball in play compared to about every 36 times last year.  But we weren't projecting anything close to what he's done this week.

On that note, we have to remind ourselves of other great starts in baseball history that are comparable.  In particular, I am drawn to the three home run opening day performances of George Bell and Karl Rhodes, Bell in 1988 and Rhodes in 1994.  For what it's worth, Bell went on to hit only 21 more the rest of the way that year and Rhodes hit - wait for it - just 5 more home runs after that strong opening day, even though he managed to finish with 269 at bats that year.

The extended essay I linked to in the first paragraph reminds us that just because a player starts hot does not mean that he's bound to have a balancing cold streak later.   It's that we just have to keep year-to-date stats in perspective and remember that one week rarely, if ever, reveals a complete change in ability.  It can happen but it would take something extraordinary.  For example, when Kerry Wood struck out 20 Astros in 1998 in a single game without walking a batter, whatever limited doubt I may have still had about his readiness for the majors was instantly removed.  A game like that is not comparable to a more common shutout where a pitcher strikes out 10-12 batters, walks 2 or 3 and gives up a lot of ground balls.  The distinction is crucial if we're to recognize genuine readiness or change in a player's ability.

As there always is in April, there's an incredibly good opportunity right now and that is the potential to exploit a player's hot or cold start in your trade discussions.  A player like Vernon Wells is a great example of one who is now much more likely to hit 20 home runs than the 15 I was projecting a week ago, not necessarily because he's proving that he's a 20+ home run player but because he's already got a head start and even if he's only a 15 home run type player the rest of the way, he's already got four in the books.  Of course, when you're trading fantasy players, your focus should be on the portion of the season that hasn't yet been played.

Don't be afraid to deal away players out of fear that they're actually going to have a great season.  It can happen and when it does, you pick yourself up, tell yourself you got that one wrong but played the percentages and then move forward.  What you're interested in is long run ability and what is likely.  Let's continue with Vernon Wells.  He's thirty-one years old and to give him credit, he's been an exceptionally good sport given all the heat he's taken for a massive contract that's now in the $20+ million per year range, obviously anxious to prove his critics wrong.  But at his best in his career so far, he's had two seasons with more than 30 home runs and when he was in those best years, he was ranging between 23-33 before dropping back to the lower levels of 2007-2009, a great deal attributable to injury-shortened seasons.  Last year, he hit only 15 in 630 at bats.

Could Wells return to 30+ territory?  Sure.  Could he have a career year and hit 35-40 home runs?  Maybe. But in making fantasy transaction decisions for any player, consider the player's long run ability, the kind of ability you would bet on being the average if you played a billion seasons and averaged the outcomes of each.  If after analyzing the evidence of Wells' track record, including this excellent first week, you conclude that he's more likely to hit 15-20 home runs the rest of the way, well then now is the time to deal him if he's on your team.  Someone out there will salivate at the home runs so far and conclude that he's back to being what he was back in 2006 or maybe even better.  Perhaps he is but it's not a question of what he actually ends up doing as much as what he's likely to do that should motivate and inform our trading decisions, especially at this time of year.

In short, while it's possible that hot and cold starts reveal that a player is a completely different one than the one we saw last year, you have to remember that we're talking about one week of play when compared to entire still-recent full seasons of outcomes.  I don't remotely dispute that what a player has done more recently is a better indicator of his ability than what he did a long time ago but the great unknown is how much weight to apply to each.

On that note, let's consider the actual weight a little more.  Vernon Wells, as of this writing, had 17 at bats in the books this year.  Last year, he had 630 at bats.  Imagine that a player's real ability is comparable to a giant drum of marbles in which different colored marbles represent a different outcome.  Throughout his career, the percentage of home run marbles in the drum is constantly changing as he gains experience, ages and of course, ultimately declines.  Now, last year we drew 630 times from the drum labeled "Vernon Wells" and pulled 15 home run marbles.  This year, we've pulled 17 times and have already pulled 4 home runs.  How much faith are you going to put in those 17 recent draws and how much will you put in last year's 630 draws?  Can we conclude that the marbles in this year's drum are made up of an entirely different proportion than last year's based on those 17 draws made this week?  Ultimately, each of us has to decide this for ourself in the context of our comfort level and confidence.  I personally try to do this as we adjust our forecasts throughout the season and certainly, by the third or fourth week of April, a player's start does begin to significantly affect the forecasts.

For now, you've got an opportunity to account for an unexpected start without over-weighing it in your trade discussions.  Above all, never be afraid to get individual decisions wrong as long as you're always thinking about the long run and your average outcomes.  I've seen players lose their league because they were afraid that a player was about to have a career year and thus, didn't want to trade for him after a hot start.  Perhaps they fear acquiring a player like Javier Vazquez, who just had one of his worst starts in a couple of years, because they think one terrible start represents something wrong with the player.  Sure, it can happen.  If you traded for Chien-Ming Wang after last year's miserable beginning to the season, you got burned even more after he never recovered from it.

But remember that fantasy baseball and winning is about playing the percentages and accepting that by the nature of that, there will be times you get it wrong.  The key is to always be putting the odds in your favor and taking advantage of the far too heavy weight that some others put into April stats.  No doubt, some players really will be revealing entirely new abilities this month.  In the long run, though, when it comes to veterans with strong job security, the first week is just another week, April is just another month and Vernon Wells is not likely to become Ryan Howard.

Wednesday, April 07, 2010

Desmond and Projected Disappointments

I received a somewhat strongly-worded but still worthy mailbag question the other day that I'd like to address in today's entry:

"Regarding your forecast for Ian Desmond, I think you're nuts.  Guzman can't physically play shortstop anymore and the Nationals have nothing else to lose by giving Desmond the at bats.  I think you're way low on his ABs and production.  I'd recommend researching Guzman's shoulder issue a bit more, as you'll find he can't make the throws anymore."

Now readers know that it's very rare for me in recent years to give publication space to questions that come across this way as it just encourages a harsher tone (usually comments that call me 'nuts' never even make it to my pre-screened inbox) but this one happened to make it through and despite the way it's phrased, I think the concerns raised are good ones that are worth discussing.

The first clarification that's needed is the one that we had already published in the notes included with the opening day projections, just in case some readers missed it:

"We are aware that Washington has named Ian Desmond the starting shortstop. A close look at our forecast will reveal that we're not forecasting him to hit well enough over the long run to hold down the job for the whole season (our current forecast is for a batting average around .240). The same could be said in the American League for Oakland's Cliff Pennington, currrently projected to hit in the low .230s. If they outperform our ability forecasts, both of these players will likely exceed our projected playing time as the two are tied together."

I think the reader already understood this and the contentious issue here seems to be that even if Desmond performs as projected, the reader believes that he will continue to get playing time regardless.

On this note, I want to clarify further even beyond the remark listed in the projection notes.  It's important to understand this in terms of all the forecasts and not just Desmond's.  Just because we forecast a player to hit only .240 doesn't mean he will hit exactly .240.  In other words, it's entirely believable to me that Desmond could hit .241, as projected, and still keep his job beyond the 231 at bats we're forecasting.  Rather, there's a range of possible performances being accounted for here and if .241 is the average outcome, if you could imagine playing the season a billion times, there's a wide range of possible outcomes. 

In other words, a player projected to hit .241 has a decent chance to be hitting .220 or even .200 after 200+ at bats.  If that happens, no matter how good he is projected to be or how much Washington or any team would want him to learn, he will not keep his big league starting job.  In fact, if he's hitting .150 after 60 at bats, he will lose his job extremely early.  Albert Pujols couldn't lose his job that early.  Ian Desmond could.   Distinguishing between those types of players is part of the challenge the forecaster faces and one of his most important tasks.  To address the reader's comment, there actually is something to be lost by playing such a player after he struggles so miserably as you can ruin him permanently as he attempts to overcompensate for his struggles by constantly grasping for any change that will make him ready quicker.

The second part of the reader's email suggesting I should research Guzman's shoulder more was a bit strongly-worded in that it implied that somehow I don't follow aging or injury concerns closely.  But let's go ahead and get into that.  Just over a week ago, the Washington Post carried an article that interviewed Jim Riggleman about the decision to go with Desmond over Guzman at shortstop.  In the article, which you can read here, Riggleman was quoted as saying that Guzman's shoulder was not a factor in the decision and then went on to say of Desmond that "...he may not be playing good in May, so Guzman may be our shortstop.  To open the season, we're going to give Dessie a shot there to hold that position down.  We hope that works."  There are a lot of reasons Riggleman might insist Guzman can still play shortstop, perhaps to maintain his trade value, but this didn't read to me like a manager unwilling to go with Guzman if the Desmond experiment doesn't work out.

Our Guzman forecast isn't exactly lights out either. The moment Desmond moved ahead of Guzman in mid-spring, we downgraded Guzman significantly until he ultimately landed at just over 400 at bats in our opening day set. This isn't all coming at shortstop and Guzman will be fighting early for utility at bats and pinch-hitting appearances.

Just as important to understand here is that setting aside Guzman for a moment, teams find someone else to play if their top prospect rookie doesn't work out.  It doesn't have to be Guzman.  It can be a player claimed off waivers in May.  It can be someone else who breaks through early in the year in the minors.  It's not about whether there are viable options when it comes to not ruining a prospect.  They get someone else to play there if the prospect doesn't pan out.

So, if Desmond does have one of those 10-20% type outcomes that lands him in the lower .200s, he will lose his starting job and someone else will be playing shortstop in June or July.  Now Desmond may not end up disappointing and as I say, it's possible that if he lands precisely on our forecasted ability, that he keeps his job.  But if our forecast is at or near the midpoint of expectations, as we often describe it, there's a lot of room on that "under" mark of our forecast to end up having Desmond back in the minors.

Today's entry really isn't about one player.  It's about the idea that when we consider projected playing time, that forecasted current ability is very much tied to the forecast.  As I often say, if we prove to be outright wrong about a player's ability, particularly underestimating a prospect's readiness for the majors, then we will definitely be wrong about his playing time.

On that note, as for whether we're underestimating Desmond specifically, yes he had a strong spring (.318, 1 HR, 13 R, 15 RBI, 6 SB in 68 AB) and a very good 2009 minor league campaign but as recently as 2008, he hit just .251 at Double-A and the year before that, .264 at High-A and the year before that, .244 at that level.  The great unknown here is how much weight to put into what is more recent versus overall.  He's still relatively young (24) but for now, looking beyond last year's strong Triple-A performance in 55 games, here's how he's done in his career so far at each level:

AA .252 in 614 AB
High-A .255 in in 1,042 AB
A and Low-A .247 in 308 AB
Rookie Ball .236 in 229 AB

It's the above numbers that have us concerned as none are even reduced to account for the inferior quality of competition at each level.  In making our forecast, we have accounted for that plus the effects his experience should have on his improvement, his more recent strong performances and so on.

In short, I wanted to highlight Desmond today as just one example of many.  Clearly, so much of our projected performance is directly tied to whether we prove right about a player's unreadiness for the majors and I completely understand the frustration of readers who want to see us go along with what they expect from a player.  A player may very well end up being exactly what the reader imagines and I have no problem conceding that.  In other words, if Desmond is hitting .270 in June, we are going to be wrong about our forecasted at bats unless he gets hurt.

In fact, while I want our projections to be as accurate as possible, I really do root for up and coming prospects as I always enjoy seeing the next generation of players break through.  I'll be the first to admit that especially with players with such limited big league data, there will be a high percentage that we're plain wrong about.  Whether Desmond specifically proves to be one of those isn't the issue here.  It's that I wanted to offer reassurance that we do consider everything possible when putting together a forecast and we're not simply ignoring the apparent plans of each team or the health status of each player when we seemingly go against consensus projections.

Saturday, April 03, 2010

Opening Day Commentary

It's become a tradition for me to publish a commentary just before opening day each year and this season will be no exception.  To varying degrees, much of the material I publish throughout the year in this space is either rooted in some sort of scientific method, whether it be completely unproven theories (such as our recently-discussed Asymmetrical Auction Strategy talked about this week) or more well-established methods, like our constantly revised forecasting process for individual player forecasts.

Today's piece is the exception.

In other words, what we're about to offer today isn't a scientific projection.  While I can't deny that my sense of each team is certainly influenced by the individual player projections we publish, we've often reminded readers that our focus on the individual forecasts does not enable the reader to add up the sum of the parts to make accurate estimates of how a team will do.  For example, we haven't forecasted every single at bat or inning for each team and there is an unfortunate but deliberate gap between how many innings a team will actually get and how many are projected in the individual forecasts.  This gap represents the complete unknown, players who are starting the season at Double-A and who get added to the 40-man roster in May and called up in July when three pitchers go down with an injury the same week.  For some teams, like Kansas City, this gap is much wider than, say, the one you'd find for the Yankees.

So, as I say every year, there's no need to send me an email to prove me wrong here as I'll go ahead and say that these are best guesses, my very rough approximation of how I see each of the division races shaping up.  We have no doubt that many readers are capable of forecasting the races much more accurately than I could ever pretend to do.  I even contemplated discontinuing this tradition a few years ago until readers told me that they can accept the warnings and the crudeness of the whole operation as if we were having a casual discussion and I was just openly saying how I thought each pennant race might go.  So, if you're still reading and are prepared to tolerate that, here's my take on the 2010 season:

American League East

1. Boston
2. New York (projected Wild Card team)
3. Tampa Bay
4. Toronto
5. Baltimore

I think both Boston and New York have incredibly good teams and with that said, the only surprise in this division will be ranking Toronto ahead of Baltimore.  No, it's not because I root for the Blue Jays and this ranking is almost a coin toss between the two.  I think Baltimore's pitching is going to be disastrous this year, among the worst in the majors.  As for the top of the division, I believe the Red Sox have put together the best run production machine in the majors and their pitching is comparable to the Yankees.  Most readers know that I have usually picked the Yankees in recent years but this year is an exception.

American League Central

1. Chicago
2. Minnesota
3. Cleveland
4. Kansas City
5. Detroit

Picking Detroit last here will not be popular with my friends in Michigan and this isn't so much a forecast to disastrous results as much as it is as I think the Royals are about to improve, especially on the pitching side, edging out Detroit.  I used Kansas City as an example in the introduction about the great unknown, those players we can't forecast who get called up later in the year when players go down and that will very much be a factor here.  I'm expecting Detroit's rotation to greatly disappoint compared to widely-held expectations.

American League West

1. Texas
2. Los Angeles
3. Oakland
4. Seattle

Yes, I considered Seattle's pitching and defense and I'm not picking them last to be a contrarian.  I think they're going to have great difficulty scoring runs and will be among the lowest-scoring teams in the majors.  I believe this is only the second time ever that I have picked Texas to win the division but it was very close and I could see a battle between the Rangers and Angels going right down to the final days of the season.

National League East

1. Philadelphia
2. Atlanta
3. Florida
4. New York
5. Washington

The gap between Atlanta and Florida is quite wide here and the Phillies remain the class of the league.  In fact, they are arguably an even better team than last year.

National League Central

1. St. Louis
2. Chicago
3. Milwaukee
4. Cincinnati
5. Houston
6. Pittsburgh

I won't be surprised it the Cardinals and Cubs are fighting it out in this division until late in the season with Milwaukee looking in from a slight distance.  I doubt the projected order in the bottom half here will be perceived as even remotely contentious.

National League West

1. Arizona
2. Colorado (projected Wild Card team)
3. San Francisco
4. Los Angeles
5. San Diego

If I'm right about this one, it will be because Arizona delivers what I think is going to be a surprisingly good offense and at least average pitching.  The Dodgers slide well back here because, like with the Mariners in the American League, I don't believe they have the run production ability to match the pitching staff and while there are unquestionably a few stars here, it's the overall offensive output that I'm questioning.

Overall

In short, I'm projecting disappointing seasons from the Tigers, Mariners and Dodgers with Arizona being the surprisingly good team of the 2010 season.  I never try to guess award winners but if we're looking at how the post-season shapes up, I expect it to be a Boston - Philadelphia World Series with Boston emerging as the best team of 2010, the Yankees (projected Wild Card winners in the AL) second and Philadelphia third.

I wish everyone an enjoyable opening day and, for those who are celebrating it this weekend, a Happy Easter. - DL

Thursday, April 01, 2010

Asymmetrical Auction Strategy II: Keepers

The other day, I blogged about a theoretical strategy designed to achieve the 78% discount target in fantasy auctions. I was pleasantly surprised at the number of readers who instantly grasped the concept despite my admitted difficulty at summarizing the idea. An instant theme emerged, though, where many readers wondered how to apply it to their keeper leagues. In fact, I think keeper leagues would work well with the strategy but the emails so far caused me to want to clarify and amend my comments of the other day.

Returning to the chart I published (if you missed the concept or want to refer to the chart, we encourage you to read the full article here), we have 34 different blocks, each representing $10 of the total goal of achieving about $340 in total projected value by spending only $260, both by the end of the auction process. The top blocks, the ones we would expect to use on superstars early in the auction, enable us to deliberately overbid for such players with an eye on picking up bargains later.

But suppose you have some great keepers and you want to account for them in your strategy, as you should? Well, there's no rule that you have to achieve these blocks in sequential order. For example, if you refer to the chart we published in part one, you'll see various discount amounts for each stage. Let's say that you have a projected $20 value keeper reserved at only $10. That means that you already succeeded at picking up an undervalued player that saved you about $10 so you can look in the chart and find two blocks (because you've got $20 in projected value) that add up to $10 in savings, such as blocks 290 and 310, which together happen to add up to a target of savings of $10.

So, you cross those two lines out, already content in the knowledge that you've got a $10 savings and $20 projected value achieved in the bottom part of the goal chart. You can do this for every keeper or your entire keeper list at once as there's enough of a scale that you should have little trouble finding a match. Have a $30 player for $18? Well, then you want to cross off three blocks (it's important to remember that each $10 in projected value achieved means another block needs to be removed) that happen to add up to $12 in savings. In this case, we could cross out blocks 230, 240 and 250, which add up to approximately $12 in savings and $30 worth of block value.

The blocks don't even need to be adjacent in the chart, either. For example, let's say you have a $20 player who only is costing you $13 on your keeper list. You could cross off blocks 190 and 240, which add up to $20 of target value (two blocks) and a target savings of about $7.

One other clarification I want to make about the idea is that you don't necessarily have to carry down savings during the auction to the first available remaining block, as I described in part one. You can instead easily distribute that savings among several remaining blocks to distribute your remaining overbidding money among several players or remaining stages of the auction. As long as you put the savings so far somewhere in your remaining chart, preferably earlier than later, by the time you finish the process you will have achieved the target value for the auction.

Tuesday, March 30, 2010

The Asymmetrical Auction Strategy

This is an automatic notification to RSS readers that today's blog entry is an extended essay only available for viewing directly at the BaseballNotebook.com website or directly through this link.

Friday, March 26, 2010

Optimal Bidding - APPLYING THE 78% RULE

Most regular readers know of my so-called "78% Rule" but if you don't, it's pretty simple and no useful advice on fantasy auction strategy can begin without understanding the general concept.

In short, the 78% rule is a guideline that says that if you end up with a team that costs about 78% of its eventual value, you're probably going to be a strong contender.  Now, there are no guarantees and just because you had a team at 78% of its projected value, it never assures you that its real eventual value will end up being the same.  For that to happen, your team needs to perform as projected, of course.

The 78% rule is rooted in this idea: The average team ends up acquiring average value at average cost.  There's no escaping this.  So, if you end up paying what a player is worth based on all of the auction dollars being assigned out, then you end up with an average team.  The goal is to be better than average.  We originally came up with the 78% number more than ten years ago when we were able to study hundreds of final league results to see what the winning teams had done to acquire players.  There was no consistent number and in some leagues, the winning team might have only had to pay 85% of eventual value on auction day salaries and in other cases, they did well and managed 70% but when the dust had settled on our review, the average winning team seemed to acquire talent at a discount that worked out to paying about 78% of actual eventual value.  The percentage continues to work well and we keep using it in our own targets.

Now, easily one of the most common questions that arrives both in my mail bag and to our member support team has to do with the surprising player values that show up for new users.  For example, we often hear something like this:

"Albert Pujols is coming out as only worth $28 in my league.  Players like Pujols typically go for $40 and up in my league so how am I going to get Pujols at 78% of that $28 value?  I wouldn't even be able to get him for the $28."

There are several telling statements in a message like this and I literally can't count the number of emails that read almost exactly like this one.

First, there is an understandable tendency for readers to want projected dollar values that are in line with what they expect to see rather than what actually will help them win.  In other words, every change in the league parameters modifies a player's projected dollar value and the further you get from what an imaginary so-called "standard league" has, the more likely you are to see projected dollar values that are different than what you are used to.  I'm going to exaggerate here to make a point.  Let's pretend that each team in your league drafted the usual number of hitters but was allowed to have only one active pitcher.  What do you think that would do to the dollar values for pitchers?  Well, there would be very few pitchers with any value at all and each would be enormously important to your chances of winning and their projected dollar values would reflect that.

Let's take it to the other extreme.  What if you were required to have an active roster of fifty hitters.  Would there be any chance that Albert Pujols would be worth $40 to your team if you have a grand total of $260 to spend?  Of course not.  In short, our ranking sheets are trying to measure the value of each player as they contribute to winning given the parameters of your league.  It's the reason why starting pitchers tend to be more highly valued by us than many people expect them to be.  They are key contributors to assembling a winning team, especially in 5x5 leagues.

Of course, and this is really the trap that we all have to be careful to avoid, there is a two-part presumption here.  First, just because someone in your league always takes the bidding up to $40 for Pujols doesn't actually make him worth $40.  Look back at last year's league winner.  How many such players did he have on your roster?  We're not saying you can't win by overpaying for star players but go ahead and have a look at last year's rosters and salaries.  Did the winning team end up with any eventual bargains, players who far outperformed their salary?  Chances are good that they had several of them, at least.  How many of those $40 type salaries were on the winning team?  One?  None?  The second part of the assumption here is that the reader seems to think that you need players like Pujols to win a fantasy league and this simply isn't true.

Returning to the 78% rule, what you want is to target assembling a team for about 78% of its eventual value.  So, in a $260 league, you're trying to end up with about $333 worth of players or more, if possible.  However, and this is the key clarification, the strategy doesn't mean you have to pay exactly 78% of each player's value.  In fact, trying to do that during an auction could easily put you in a situation where you're sitting there two thirds of the way through the auction with plenty of dollars to spend and no players.

What you're trying to do is achieve that 78% by the end of the auction.  In other words, let's say you're absolutely convinced that you do need a player like Pujols to win your league.  You can not only honor the 78% target but also bid in line with how your league values superstar salaries.  To do this, you overbid on Pujols and compensate for the overbid by targeting bargains later in the auction.

Let's continue with the example above.  Let's say that Pujols is showing up as a $28 value in your league and you know that it will cost you $41 to end up with him and you want to do that.  So, here's how the math plays out in this situation:

You started with $260 and spent $41 on Pujols.  Your target for the whole auction is $333 of player value because if you end up with $333 worth of player value for $260, you've bought a team at about 78% of its value.  So, you now have $219 left to spend (i.e. $260 minus the $41 you spent on Pujols) and you still need to acquire $305 worth of value (i.e. $333 minus the $28 estimated value of Pujols).

The implication of your significantly overpaying for Pujols is that needing $305 more value with $219 is that your target for the remainder of the auction is to pay about 72% of projected value for the remaining players or $219 to spend to acquire $305 of remaining value = 71.8%.

This is why keeper leagues are so great for applying the 78% rule.  With a tremendous keeper list, you can easily find yourself in a position where your remaining target actually goes above 100%.  This can also happen early in a draft if you end up with some bargains and your modifier will actually change throughout the draft.  You shoud keep track of the following two items constantly: How much remaining value do I want to end up with?  How much do I have left to spend?  As long as you can answer these two questions, you actually can bid with your instinct, especially in the first half of the auction.  Of course, you can't wait forever to achieve your targets and you have to know that you're rarely going to end up with $40 of projected value for only $1.

In short, this is where fantasy auction strategy really does depend on your knowledge of what it takes to win and not just a mechanical bidding method.  If you think Pujols is the type of player that you can build a winning team around, you can bid whatever you want for him without defying the 78% target rule.  In fact, I suggest that in most leagues, there are cases where you have to overpay for some of the well-established producers.  What you try to do is compensate for the overpaying by adding other potential bargain picks in the second half of the auction, the riskier picks who are projected to do well but who will be much more easily (and cheaply) acquired later.

Wednesday, March 24, 2010

Bumgarner, Thames, Marlins 1b, D. McCutchen

As spring training nears the point where clubs are starting to make fairly major decisions about how their rosters will look on opening day, there have been several transactions in the past few days that will have significantly affected the forecasts.

As we alluded to last week in this space, Madison Bumgarner did end up getting sent to the minors and it's a major disappointment for him.  No one can blame the Giants for this move, even those who don't put much weight on spring performance.  Bumgarner wasn't just ineffective this spring.  His velocity was way off, just as it was late last season even as he continued to put up good results.  As it's now official he didn't win the fifth spot in the rotation, it won't surprise me if he spends most or all of this season in the minors and I'm downgrading his forecast for 2010 to almost nothing now.  In keeper leagues with deep reserve lists, he remains a good target, of course.

Marcus Thames appears to have won a bench spot with the Yankees and while I'm not about to project him to get full-time at bats, Thames is well-proven at being able to hit for power even when in a limited role.  He strikes me as the kind of player who could find 200+ at bats here and if that proves to be the case, he's a very cheap 10-15 home runs you can add to your bench.

Jorge Jimenez, a Rule 5 pick from Boston, failed in his bid to win the Marlins first base job and was shipped back to Boston this week.  That means that Gaby Sanchez is a good bet for 300+ at bats this year barring Logan Morrison breaking through somehow.  Sanchez's only expected contribution will be the occasional home run and he could hit 15 if he lasts the full season in the lineup.  Jorge Cantu appears to be moving back to third base full-time now.

One other event that caught my eye was that the Pirates named Daniel McCutchen as their fifth starter.  While McCutchen is good enough to last the season in the rotation, we're currently forecasting an ERA in the high 4's and obviously, he wouldn't be a strong bet to get many wins with this team even if he pitched better than we're forecasting.  He doesn't project to be particularly helpful in strikeouts or WHIP either and while this is a well-earned opportunity for him, he's a lot older than most would guess, already twenty-seven as the season gets underway.

Sunday, March 21, 2010

CLE, MIN Closer Situations

Two fairly significant injuries were announced after the cutoff point for consideration for this morning's published projection set, one of which we had pretty much expected and the other being a surprise.

First, the news we didn't see coming was that Kerry Wood could be out of action until May or June with a strained lat muscle.  As it happens, we had just given next-in-line Chris Perez 4 saves in the new forecasts and Perez's saves forecasted will now be upgraded to at least 10+ now.  Wood will be downgraded from 25 saves to no more than 16 or 17, I expect.  Perez is good enough to hold the job until Wood comes back and readers should keep in mind that there is always the risk of a player like Wood not making it back as soon as expected.  Needless to say, Perez now becomes a must-own in fantasy leagues that use saves as a category.

The other news item that was pretty much expected was that Joe Nathan formally confirmed that he will undergo Tommy John surgery and miss the 2010 season.  In this case, it was a pretty likely outcome all along since the injury was first announced.  In line with that, we had already downgraded him to just 6 saves left in the version of the forecasts that went online for this morning, with that forecast now to be erased completely.  Of course, there's no guarantee who will step in to replace him and while we've projected saves for all of Jon Rauch, Matt Guerrier and to a lesser degree, Jesse Crain (we still don't believe Francisco Liriano will be moved into this role), it still wouldn't surprise us to see the Twins bring someone in from outside.

If that doesn't happen, the potential contribution from any of these players could be temporarily measured by not giving much regard to the saves.  That is to say that any one of them could easily be declared the closer to start the season and be out of that role within weeks.  So, if you're drafting anyone from the Minnesota bullpen, I'd recommend focusing on their other potential contributions to your team without thinking too much about whether the pitcher is the closer of the week.  Rauch, Guerrier and Crain all are projected to be above-average relievers this year, with Crain being projected to bounce back from last year's disappointment.   So, any one of them can round out your pitching staff by offering a slight bit of benefit to ERA and WHIP.  If you think of each pitcher that way, and you happen to end up with one of the players who doesn't end up in the closer's role for now, you still won't be overly disappointed and they may ultimately get a chance at that role later in the season anyway.

Friday, March 19, 2010

Bumgarner, DeWitt, Toronto Closer

It's time for me to openly answer some questions that have been sent to our mailbag and we'll continue to work in as many of these as possible in the coming weeks:

Q. Has Madison Bumgarner's spring cost him a spot in the rotation?

A. Yes, it's certainly in serious jeopardy anyway.  I often emphasize that you can't put too much weight into spring training statistics when it comes to forecasting ability but even if we don't do that, teams do when handing out roles on the big league squad for April.  In this case, it's understandable too as even in the small sample size of seven innings, Bumgarner has walked seven and struck out none and that's an alarming ratio even over a short stretch (to appreciate how bad that is, imagine how often you would see that sort of line in a regular season boxscore).  We weren't forecasting him to be an instant superstar, as he appeared to be in September of last year, but we didn't expect this sort of terrible spring either.  I expect he's headed to the minors to start the season unless he puts in a couple of dominating performances very soon.

Q. Who plays second for the Dodgers?

A. Those who subscribe to our extra optional spring training sets will notice I just upgraded Blake DeWitt to over 300 at bats in his latest forecast.  In other words, no matter what happens this spring, he's not outright guaranteed to get the most at bats here because being declared the second baseman in April (which he now seems poised to be) doesn't automatically get you the job in June.  So, even if Joe Torre says that DeWitt is his guy, I'm not prepared to go higher than about 380-385 at bats in my opening day forecast as he still projects as a mid .240s hitter with only occasional power, no speed and questionable defense.  In short, I could see him winning the job in March and losing it by May or June.

Q. I notice you've got a lot of saves projected for Toronto but no clear closer. Is this because you believe that they will go with a committee or you're just not sure who the closer will be?

A. You're much closer on the latter part of your question and this was sort of verified this week when Jays manager Cito Gaston openly implied that while he remained undecided, he'd rather give the job to one guy by the end of the spring than have uncertain roles handed out.  If I'm trying to handicap it right now, I would put Gregg slightly ahead of Frasor with the lefty Scott Downs (who has arguably looked the best of the three this spring) barely behind those two.  Just as I said above in relation to the Dodgers' second base position, even if the team declares an outright closer for the start of the season, it doesn't guarantee that player the role two months down the line if he doesn't do the job well.

If you're in an upcoming draft or auction and are so desperate for saves that you're searching staffs like Toronto's, it would be a good idea to hedge your bets here by doing everything you can to pick up more than one of the candidates.  Chances are good that if Frasor was owned last year, though, his previous owner will have locked him up for this year.

***

Later next week, I'm going to talk a bit about optimal bidding in keeper leagues and how I handle adjusting for the quality of my keeper list.  While I'll use 2010 examples and update my thoughts to remind about perception and never wasting unnecessary dollars just because a player is projected to do well, if you simply can't wait for that piece because of an upcoming auction, you'll definitely want to check out our long-archived essay On Paying 80% of Projected Value.

By the way, we've now assembled most of our participants for my mock auction experiment (invitations were sent out and accepted earlier this week and we're down to just two spots left as we await responses from aternates).  To expedite what is already likely to be an extended auction, to be held the weekend of March 27-28, we are going to keep the actual event private but in the week following the draft, we will not only publish results of the auction here for all to see but I will also explain how the theoretical bidding method I was using worked, what problems I detected in the approach and so on.  We had so many willing participants that I wished there had been room for more than 14 participants but an 80-team league just didn't feel right to me...

Wednesday, March 17, 2010

On Target (Josh Parks)

This is an automatic announcement to RSS readers that today's entry uses formatting that is only available at the site.  You can view the article through this link.

Thursday, March 11, 2010

Hitting Prospects Pre-2010

This is an automatic announcement to RSS readers that today's entry is an extended essay available only at the BaseballNotebook.com website or directly through this link.

Reyes, More Nathan, Challenge Reminder

It seems fairly certain now that Jose Reyes is not going to be ready to go by opening day, this according to an article on the MLB website published just a couple of hours ago.  The problem with a condition such as he's facing is that there never really is a specific timeline given for a return and he was already facing a comeback from a fairly serious extended absence and injury.  Even without the benefit of a timeline for his return, I do already plan to reduce his forecast to something closer to 120-125 games than the 145 we had previously published.

Though Joe Nathan is deferring a decision on whether to undergo season-ending surgery, it's looking increasingly unlikely that he pitches this season and in building his new forecast and revised versions for the likes of Jon Rauch, Matt Guerrier and Jesse Crain, I'm currently going with about a 75% chance that Nathan doesn't pitch in 2010, meaning he still gets a forecast but not much of one.  I still don't see Crain as the new closer here but, if nothing else, his importance to the Twins goes up and his job security as a key arm out of the bullpen also improves.  It still wouldn't surprise me for the Twins to deal for someone who's closed out games more recently than any of the current candidates.

By the way, we already have almost a hundred people who volunteered to participate in my mock auction to test a new bidding method and we will be making the draw tomorrow with emails to go out to those selected by Monday.  With so many willing participants and only 14 spots available, obviously most will be disappointed but that's the nature of picking randomly from a large list.  If you'd like to be in the draw to be involved, be sure you send your email by 11:59 PM EST tonight as per the instructions published when I first talked about the idea.

Tuesday, March 09, 2010

Nathan, R. Martin, Playing Time Changes, Prospect Week

Since this weekend's projection set was published, we learn that Joe Nathan may be facing season-ending surgery because of a torn ligament in his elbow.  The gap between Nathan and every other reliever currently under contract with Minnesota is incredibly wide and there isn't a true "future closer" here who seems ready for 2010.  Jon Rauch and/or Matt Guerrier could close for now but Nathan has been one of the best closers of the past 5+ years and a guy like him is impossible to replace.  Jesse Crain's less than steller 2009 will likely cost him consideration.  There's the chance that Nathan doesn't opt for the surgery but we won't know anything for sure immediately and this could even cause a ripple effect involving another team.  In other words, maybe the Twins go out and trade for someone else's closer and open up a closing role elsewhere for another team that has a more ready closer in waiting.

Also since this weekend's projection set was published, we learned that Russell Martin has a strained groin that could keep him out for the next six weeks.  Though his exact recovery time is unclear, the official team site has reported that Martin was unlikely to play on opening day.  We were already projecting a lower-than-usual games played total for Martin, not because we could possibly anticipate this injury, but because we believed it would be quite difficult for him to maintain the incredibly high games played totals he's been getting as a catcher, if for no other reason than he would start needing more rest than he's normally received.

By the way, I briefly want to clarify that constant modifications in our projected playing time throughout spring training are not exclusively based on events that you read about in the news.  We've hit a point close enough to the season now where we're trying to examine how a roster shapes up and how the presence of certain players impacts the playing time of others.  We don't force balance the way we used to back five years ago when we forecasted exactly 162 starts for pitchers and the right number of plate appearances for the hitting side of the ledger and so on.  We discovered that attempts to insist on so-called realistic totals actually hurt the accuracy of our individual forecasts even if it helped the appearance of team summary information.  Still, we cannot deny that certain players do clearly impact another player's playing time or job security, especially in cases where options on the bench are arguably superior to the starting player and so on.  I wanted to emphasize that because not every change we make at this stage is about a battle won or lost in the spring or the result of an injury.

Finally for today, I'm pleased to confirm that this is "prospect week" and that means you'll see the return of our dynamic prospect ranking lists, with the first hitter list for 2010 coming by Friday morning and the first pitcher list by Saturday morning.  As we did last year, we'll identify 50 players (25 hitters, 25 pitchers) along with some additional names you need to know about and this will define the baseline prospect set for the 2010 season.  Once this baseline is established, as we did last year, starting in the final week of April and through to the end of the full minor league season, we'll revisit these lists every Friday, alternating between hitters and pitchers.  We're going to try to build on what we started with this trend last year and if you weren't here last year, you can still check out the final 2009 hitters list and pitchers list, which is in a similar format to what we'll do again this year throughout the season.