Ask Rotoman

Evan Gallahou is a standup coming looking for gigs as a fantasy baseball auctioneer. I’m not kidding!

The following clip starts with a baseball joke, but it was the bit about the weather broadcaster at the end that made me think Evan might be a really good auctioneer.

The movie of Fantasyland, Sam Walker’s book about his season as a fantasy rookie taking on the Tout Wars experts league, is coming out on Friday at www.snagfilms.com. I got an early chance to see it and here are some thoughts.

I wasn’t a part of the Tout LLC when the movie was made, so I don’t have any behind the scenes info. Jed Latkin, a trader in New York, was chosen as the regular guy to join Tout AL (which included Sam Walker) for the 2008 season. The movie takes a general approach to setting up the fantasy game, including graphics with statistics indicating the bigness of the fantasy sports world, and interviews with Jed and others who applied for the regular guy role, in which they brag about their charmless obsessiveness. Some of this can be ascribed to their competing for the job, but it isn’t pretty.

Jed, somehow, is the most charmless of all (he has a line where he says that with his child scheduled to be born on draft day that being an involved parent doesn’t mean being there for the birth, necessarily), but his alpha trading personna is actually winning as a character. Perhaps that’s why the filmmakers focus almost exclusively on his exploits, and ignore most of the players in the league. Jed also has the perfect sidekick, his patient and generous wife, who softens his rough edges. If she can put up with him, well, maybe we can, too.

Of the other 11 members of Tout AL, the film makes Ron Shandler the big dog, the guy Jed wants to beat. Ron plays the part to perfection, adopting a “what’s that, a fly?” attitude toward Jed, disdainfully criticizing him for trying to make a trade in April, and for driving down to Roanoke Virginia to Ron’s house, without calling, to try to seal a deal. These scenes are funny, and made me wish there was more interaction between Jed and the other combatants. The film feels, in terms of storytelling, a little thin, but the set pieces (Jed goes to Spring Training to introduce himself to his guys: “Hey, Justin Verlander, welcome to the Jedi Knights”) are funny, and as we go along the movie’s focus on Jed’s voice pays off.

It turns out that this is a story of one man’s attempt to live his dream, and really all we need to see of it is his point of view. His reactions and attitude are strong enough to open the window into all fantasy players’ psyches, at least partway. Fantasyland, the Documentary, isn’t great filmmaking, but it is competently done and should be of interest to everyone who has played the game that obsesses me. That it is a bit pathology as well as entertainment might be uncomfortable, but it isn’t a bad thing.

There are many clips and outtakes at YouTube, which will give you an idea of the film’s flavor.

The Book Blog

Tom Tango looks at some assumptions we make about forecasting player performance, and looks at the race for 2009 HR champ to illustrate how the high variability of performance means that our forecasts–even if we show just one number–really outline a range of scenarios. In the optimistic one, perhaps Albert Pujols plays the entire season, faces weaker than usual pitchers, and matches the scenario at the top of the forecaster’s scale.

In the most pessimistic scenario, Pujols throws his back out the day after the forecast is made, and doesn’t play all season long. So, if he might hit 50 at the top end and 0 at the bottom end, what is a good projection for Pujols?

I’m not a fan of assigning percentiles of probability, as PECOTA does, because they don’t really mean anything real. From the comments on this post of Tom’s I learned that it seems that PECOTA applies the same distribution rules to all players, which may match the knowledge we have now, but certainly doesn’t give us any more information.

To make my projections I’ve run regressions that give me a baseline formula for using the information available in a player’s past performance, modified mostly by age. The problem with this approach is that the regression uses absorbs the volatility of the sample and spreads it throughout. So, if I apply the formula to the Top 100 projected hitters, I get a projected number of at bats (and other stats) about 10 percent less than the Top 100 hitters produced the preceding year. That loss can be attributed mostly to injuries, since these are generally reliable players.

The problem is that applying this loss across the board makes all the projections look weak. No player gets 600 AB, nobody hits 40 homers, things just look wrong. This is exactly what Tom Tango’s “dumb” projection system, Marcel the Monkey, does. Marcel only looks at previous stats and age and applies its regression formula. This gives an excellent projection of where production will come from and how much production will come, but while it tests well it doesn’t look right.

Projections are pretty limited in their applicability to allied uses, like team forecasts, but they are a good way to present the information about what is expected of a player. Does he run? Does he hit for power? The projection aggregates the information we have about a player and comes up with a compromise view that helps us smooth over the ups and downs of individual statlines. But to make it look right, you have to add that 10 percent of at bats and stats back into the individual lines, even though this means projecting too much stats overall.

Nowhere did this become evident more quickly than in the chart Tom ran in his post showing Marcel’s top 13 HR projections for 2009. In the first column are the Marcel forecast HR for each player. In the second column is the player’s name. In the third column are my 2009 projected homers. In the fourth column are my projected homer totals reduced by 10 percent. In the fifth column are the actual total homers.

As Tom concludes, we can get the right number for the group. The real question is what do we want the projection to do? The post is well worth reading, as are the comments, if you’re interested in this murky side of the sabremetrics game.

Everybody knows about Joe Nathan, the man with the most saves in baseball the last six years, who has a tear in his ulnar collateral ligament. The problem right now, for me, is that I’m preparing updated projections for the Patton Software and there is no way to know whether Nathan is out for two months or two years. You see, the odds of Nathan getting back onto a field after rehabbing from surgery are real long, so the first medical approach is to wait a couple/few weeks, try to strengthen the supporting muscles, and see if he can pitch through it.

Not many do, but if he can, then he might get a few months of the season in and have some value this year. If he can’t, he has no value this year at all, and no value next year either. So, what should I do with his projection? And what should I do with the interesting set of relievers in the Twins’ bullpen, any of whom might actually be able to do the job if given the chance?

Let’s call what I do “pussyfootin’,” because it’s a lot like the gait of Violet, the cat that just walked over my keyboard and curled up on the back of my desk and didn’t knock over a thing (and only introduced a few typos along the way). I’m careful, thoughtful, and when I’m clear I leap. And, like Woody Allen, I always usually (yeah, right) land on my feet.

In the new set of projections I cut Nathan’s projection in half, to 35 innings pitched, and I bump his ERA and WHIP up just a bit, then cut his bid price down to $10, which I don’t think I’d pay if I was drafting tonight, but I do think someone else would bid $11 if I did. $10 isn’t likely to be the bid price in two weeks, when we’re supposed to know more, but it does reflect the market now. I don’t think you want Nathan, but you don’t want someone else to get him too cheaply. There is too much we don’t know.

My first impulse after Nathan’s injury was to bump up Jon Rauch’s projection, giving him most of the saves, but while I still think he has to be the favorite to win the job, because he did some time in the past as a closer in Washington, he’s not a lock. I made him $9 at first, because he can earn that as a middle reliever even if he doesn’t get the job, but I’ve now knocked him down to $7 because, well, there are too many alternatives to assume that he will get the job, and too many questions about his work last year to be confident he’ll hold the job if it is given to him.

Matt Guerrier is usually cited as next in line after Rauch, but even though he was a closer in college and has been an excellent middle reliever–other than in 2008–he doesn’t profile as a closer. He doesn’t blow guys away, in other words. I’ve bumped him up to $4 (he earned $15 in 2009) and allocated him some of the saves sliced from Nathan’s line. I think that’s safe, even though he doesn’t have closer upside.

The guy everybody likes for the job, talentwise, is Pat Neshek, who missed most of the last two years following a 2008 breakdown that led to TJ surgery. He’s healthy now, but still working his way back. He’s got an interesting sidearmed delivery that is deceptive and brings lots of movement. Historically, he hasn’t had much of a platoon split. The issue is whether he is really back. Chances are the Twins aren’t going to throw him into the fire immediately, so I give him a few of the saves and a bid of a couple dollars in the new version. You have to be aware of him, but he’s still a long shot at this point.

The other closer-quality pitcher on the Twins’ staff is Jose Mijares, who is the only lefty in the Twins’ pen right now. Even if that situation persists he could get some saves, but he won’t get a lot of saves. I added a couple of saves to his projection, but kept him as a $1 bid. He won’t go for more until the Twins add a lefty to their pen.

Saves are a tricky business. Any pitcher going good can get saves, but we can see with our own eyes that not everyone is able to keep going good when the pressure rises. There are some who say that Mijares is a choker, but his Leverage Index (see baseball-reference.com) shows that he performed best in the toughest situations last year. Until we’ve seen a big enough sample, it’s impossible to really judge a pitcher’s readiness for the role, but easy to understand why guys in high leverage jobs lose their jobs before they can prove that they are victims of the random thing.

Projecting player performance is a tricky business. The talent evaluation part is fairly straight forward, but projecting playing time is usually the difference between a good and bad projection. While pussyfootin’, I try to split the difference, to balance the expression of talent with the possibility of opportunity. Those of us drafting next week are going to have to make catlike choices when it comes to selecting the Twins’ closer. My adjusted bid prices are an attempt to equalize the odds of success vs. price for each player.

Ps. There was speculation today that the Twins might move Francisco Liriano to the pen, maybe even the closer spot, given their situation. This isn’t an obvious move, but if Liriano is struggling as a starter it seems like a natural next step. That, of course, screws all the values above, all of which will be updated next time no matter what happens.

 You Go First: Three Questions

The link takes you to San Diego Padres’ special assistant for baseball Paul DePodesta’a blog. And it’s like sitting in a press conference with the guy who makes the decisions (though he also has a boss). Maybe other teams are as transparent, but I don’t know that. Enjoy.

I am a big fan of John Burnson’s Heater Magazine, a weekly pdf of baseball stats and analysis that makes the Sports Weakly baseball stats pages look like the Weekly Reader.

John sent me a copy of his annual book, The Graphical Player, in January, when it came out. I glanced at it then, but I was busy and it ended up on a shelf and I didn’t write about it then, which is too bad. Like Heater Magazine, the Graphical Player is crammed full of information. John is evolving a set of graphical rules for presenting data that makes it increasingly useful and understandable, and helps put a player’s skills in the context of his team and of the game as a whole.

This is not a book to use to look up a fact, though there are plenty of those in here. This is a book to browse through, to hunt for patterns in, to savor as a baseball fan the way a gourmand might taste a sauce. The good news, even at this somewhat frantic moment, is that much of the information in the Graphical Player will still pertain after the season starts. If you want to see if a player has historically been a slow starter, this book has graphs that show that he has been or hasn’t. Once you get used to the way the information is presented, this sort of research is a pleasure. The data and its context are presented as a picture.

Other features of note: John asked three writers who follow prospects to name their 60 top rookies for this year. He has compiled their rankings and notes for these 111 ROY-eligible players, with their stats (presented in a very useful format) for the last three years. This is a very helpful survey of this year’s top prospects, though it does omit my decidedly dark horse candidate Thomas Neale (who didn’t make The Guide, which shows just how dark a horse Neale is).

I also think, as documentary, that the team profile pages in the back of the book are full of useful information. They won’t surprise readers of Heater, but as with much of the book, once you get past the sheer data density you’ll be surprised how satisfying it is to see a chart of who played what position the most each month for each team. And the charts that compare each team’s production in different categories to the league average spark only ideas thus far, but clearly they help us understand what was going on. This is a new way to experience this data, and an invigorating one.

I’ve only scratched the surface of the types of information included in the Graphical Player. Some is of help analyzing baseball, while other stuff is geared totally to fantasy players. I don’t want to be grandiose, but it is an amazing accomplishment.

UPDATE: So I posted the above glowing review only to find out that the only copy of the book you can buy at Amazon currently costs $91. It’s worth every penny, of course, but that’s a little steep. It seems the Graphical Player is also sold out at Acta Publishing, the company that published it. Barnes and Noble doesn’t have it. I’ll tell you what, I’ll sell my copy to the first bidder for $75. And in the meantime, I hope this means that John Burnson sold out his print run and made a small fortune.

from Jimmy Guterman

The Clash album Sandinista! is a big sprawling three-record set that sounds like it could have been made by six or ten bands, which it was in some way. What happened was that the band wanted out of their contract with Columbia records. They saw that they were obligated to deliver three more records before they would be free, and someone had the smart idea to deliver all three at the same time. The sessions include all sorts of guest artists and performances by people in the Clash circle, with songs in a great many styles (some of which don’t really qualify at songs at all) and genres. The record never fails to charm, I don’t think, but some of it sounds like attention was flagging. That may be the dub influence. In the end, the only problem was that the record company counted the three-record set as one release, and so the band moved on to Combat Rock and Cut the Crap in pursuit of freedom, records that have their moments but which lack the epic generous delight of Sandinista!

Jimmy Guterman had the idea of remaking Sandinista! with different bands and artists each tackling a song, something of a tribute album, but to a record rather (despite the line on the cover) to the band. He called it the Sandinista! Project, and somehow managed to record covers of all of the umpteen songs by artists you might have heard of and other you have not. The result is delightful. I had been playing Sandinista! on my iPod last summer when I learned of TSP. Guterman released the mp3s of the tunes for free on Joe Strummer’s birthday. Soon I had both sets of songs intermingling amiably in the mix. The newer recordings often have strikingly different but completely agreeable arrangements, sometimes shifting genres or emphasis, nearly always hitting the mark the Clash set in the first place.

I bring this up now because Guterman is offering the Sandinista! Project for free download for the next few days at the link above. Highly recommended.

You can read more about the project at its blog.

Last Player Picked

An excellent summary of shallow mixed league strategy in general, that has very useful information that applies to the Yahoo platform in particular. If you play in this format you have to read this.

Sunday night we held the seventh annual Rotoman’s Regulars League draft. The league is a 20 team Yahoo 5×5 league. Rosters are 20 deep (four reserves), so 400 players are taken. This is a very tough league with very smart, tough competition, both in the draft and all season long. And the all season long part is crucial. Though the league has weekly waiver claims it has daily ups and downs, so maximizing one’s reserve list and streaming players and pitchers on off days is essential. I made a respectable showing the first year, but each subsequent year things got worse until I decided to take a break. I hate sucking. This year I decided to return. I missed the guys and in spite of my suckiness at it, I like the format a lot.

How’d I do? Sucky. Here’s the team:

Here’s what happened:

1. Miguel Cabrera (1B): The seventh pick overall comes after the big boys, but before you can legitimately go after someone like Mauer or Lincecum. I mean, you could, but it doesn’t feel right. So, I went after the guy I think is the best of the big boys who isn’t a big boy. He’s the right age, he’ll be helped having Johnny Damon hitting ahead of him (but maybe not Austin Jackson), and he has a lot to live down after last year’s disgraceful exit.

2. Pablo Sandoval (3B): Waiting 26 picks for a second guy is frustrating. All the obvious names went off the board. I didn’t want an outfielder and I don’t trust Mark Reynolds or Ben Zobrist at this point, and I had a first baseman already. So, it was Sandoval for me. He’s young, so maybe there’s upside, but he plays on a crappy team offensively in a bad ballpark for hitting, so he’s risky, too.

3. Brian Roberts (2B): I was glad he was around. I wanted a middle infielder who ran. What I didn’t want was a guy who’d had his first workout of the spring hours before because of a bad disk in his back. I’d read about his problems earlier in February, but somehow missed the reports of escalating malady. If I’d known I would have taken the aging statesman, Derek Jeter. Roberts says he’ll be okay, so there’s that, but players aren’t doctors. My fingers are crossed. And I took a 2B in the reserve rounds, just in case.

4. Denard Span (OF): Okay, time for an outfielder, because there were no appropriate shortstops or corners. I added three to my queue: Andre Ethier, Andrew McCutcheon, and Denard Span. All three were available as my turn approached, but then they went down on order: McCutcheon, Ethier, and–on my turn–Span. The guy I missed was Hunter Pence, who I like a tick better, but the reality is that I like Span more than most, and I got him. Continue Reading »