Sunday, April 27, 2008

Confirmation bias

From BPro's Kevin Goldstein:
Well sometimes prospects just explode, and small sample sizes sometimes are damned, as here is one player from each of the full-season leagues who is exploding, but also has the scouting reports or existing potential to confidently up their stock.
Wrong. Wrong. Wrong.

Wrong, wrong, wrong, wrong, wrong, wrong, wrong.

This is classic confirmation bias. "I thought these guys were going to be good, and, hey, they are good." Small samples are always small samples. They do not add anything to the discussion.* These guys may still be excellent prospects, but that has nothing to do with their performance in a limited set of games and everything to do with the fact that they were already great prospects. This mindset has absolutely got to be stricken from baseball coverage.

* OK, technically they add a small sliver of evidence. If a prospect was 90% likely to be awesome, the small sample tells us that he is now 90.01% likely to be awesome.** If you want to hitch your analysis wagon to that, that's your business.

** Numbers are totally made up for example purposes. You get the point, I hope.

Saturday, April 26, 2008

Bullpen strategy

Some thoughts on baseball, having just returned from Jacobs Field (yeah, you heard me) in Cleveland where I watched my Yankees lose 4-3 to the Tribe and a one out single on the bottom of the ninth:
  • If you go to a ballpark, and they offer you a choice of a "hot dog" and a "kosher dog" it is incumbent upon you as a lover of the baseball experience to choose the "kosher dog." It will be a much higher quality dog and better cooked too, not one of those pale, limp pieces of garbage you get at the normal concession stands. If this is not the case at your local ball park, you should consider some sort of protest. Furthermore, if you happen to be eating your dog in Cleveland, you get to enjoy this superior kosher dog with their stadium mustard, which kicks the crap out of the normal yellow stuff. Simply put, if you are not having this experience with your hot dog at a ball game, it simply means that you are trying to fill the hot dog sized hole in your heart with cheap imitations. I had two. This may not have been enough.
  • The Jake is a great place to watch a baseball game. The park was built with a vertical emphasis, which is how stadiums should be built. It keeps fans close to the action. Tiger Stadium was like this, and unfortunately Comerica Park is not. Between this and the hot dogs, I may start watching the majority of my baseball in Cleveland.
  • Finally, the real topic of this post. This one is real simple. In fact, it's so simple that it blows my mind that every MLB manager does not understand it. It's like this:


    For emphasis:


    You see, when the game is tied in the bottom of ninth inning, giving up any runs means losing. Always. No exceptions. If you give up a run you will not win. You will lose. It cannot happen any other way. However, many managers prefer to not deploy their best pitcher in these circumstances, preferring to use their closer after they have acquired a lead. There are only two completely fatal, totally obvious, elementary flaws to this thinking.

    First, you may never take the lead, losing with your best pitcher unused. Awesome. Second, you may take a large lead, rendering the use of your best pitcher meaningless.

    Today, in Cleveland, Joe Girardi, like his predecessor, opted to use an inferior relief pitcher, Ross Ohlendorf, instead of Mariano Rivera (or, I should add, Joba Chamberlain, who may have been being rested for other reasons) in the bottom of the ninth in a tie game and predictably lost.

    Let's go over this one more time: if Ross Ohlendorf surrenders even a single run, YOU LOSE. YOU MOTHERHUMPING LOSE. YOU CANNOT WIN. The only reason that you would not use your best available pitcher in the scenario is if you think that using him now will cause him to be unavailable for a more critical situation (which really doesn't exist anyway) in subsequent days. Mariano Rivera can pitch on back to back days. Sometimes, he can pitch three in a row. He was rested. By not using him in this situation, you are essentially betting that you can't use him tonight because you might need him two or three days in a row in higher leverage situations (which really don't exist anyway) immediately after this. This is a nearly impossible bet.

    I just can't get over this. Necessarily, run prevention is more important in the tie game in the bottom of the ninth than in any subsequent inning in which you have the lead. This is fact. This is not speculation. It is a mathematical necessity. The laws that govern the entire universe would have to disintegrate for this to not be true.

    Pitchers exist to prevent runs. They exist for no other purpose. Therefore, the best pitcher should always be used to prevent the most critical run. And no run is more critical than the run that guarantees a loss.

Thursday, April 24, 2008

Tampa Bay attendance

Over on his blog, Shysterball, Craig Calcaterra notes that the Devil Rays are having a hard time attracting fans:
The Rays are again playing games at Disney World as part of an initiative "designed to bolster fan support in the Orlando area." Last night's attendance: 8,989. You're doing it wrong, Tampa Bay.
Naturally, for those of you who know something about lolcats, after I read that I had to create this:

Wednesday, April 23, 2008

Please do not think this way

Via Tyler Kepner's blog for the NY Times (hat tip to River Ave Blues):
Speaking of Chamberlain, here’s Johnny Damon’s take on his role. It seems to be the majority opinion of the veterans in the clubhouse: “Joba as a starter, he has a chance to help us out once every five days. Him coming in and bridging the gap to Mariano, he’s got a chance to do that three or four times during those five games."
Two things are egregiously wrong with this viewpoint.
  1. The "chances" that Johnny talks about are not even remotely equal. The best starting pitchers in baseball game exert vastly more influence on the outcome of a game than the best relievers. In the last few years, only the top five to ten relievers are worth more than five additional wins above replacement over the course of the year. The top 30 starting pitchers are always worth more than that. The top five to ten starters are usually worth over seven wins above replacement. Joba has the ability to be a top five starting pitcher.
  2. There is no way on God's green Earth that Joba can pitch out of the bullpen four times in five days with regularity. Hell, even pitching him out of the bullpen three times every five days is borderline reckless. As a reliever Joba will have an opportunity to make a difference in at most half of the Yankees' games, not the 60 to 80 percent that Johnny thinks it is.
Remember, the issue is not that Joba is guaranteed to be more valuable in the rotation, it's that he has the potential to be more valuable. No team in their right mind does not first try their stud, ace level, pitching prospect as a starter. If, after two or three years, he is unable to make that cut it in the rotation, then put him back in the bullpen.

Saturday, April 19, 2008


If you're a Yankee fan, you really ought to be conditioned enough to the ups and downs of the baseball season to withstand some bumps. Pete Abraham thinks so, and this post socks it to the crazies pretty hard.

Try, just try, to be a little cognizant of the situation your team has been in so far. Some of the comments on the blog tonight appeared to be written by 8-year-olds after sucking down three Mountain Dews.

Brian Cashman didn’t promise you a wire-to-wire joyride. There’s a plan in place and there are going to be some bumps.

Brew some green tea, take a nice sip, put some jazz on your iPod and calm down. It’s April 19.

Thanks, Pete. Those of us who aren't crazy appreciate the support.

Friday, April 18, 2008

The new wave of player contracts

According to ESPN, the Devil Rays have locked up Evan Longoria for six years. Or seven. Or nine, if they feel like it.
Rookie third baseman Evan Longoria and the Tampa Bay Rays agreed Friday to a $17.5 million, six-year contract, a deal that could be worth up to $44 million over nine seasons.
Essentially, Longoria gets $17.5 million for the six years that he'll be under control of Tampa Bay anyway. At that point, he'll be 28, and the Rays can pick up an option for his first free agent year. The following year, he'll be 29 and the Rays can essentially lock him up for the remainder of his prime, the next two years, bringing the total value to $44 million.

This is an awesome deal for both sides, and we've been seeing more and more of these recently. Teams have figured out that a couple million dollars a year is chicken scratch to them but means an enormous amount to a player who is still not yet eligible for arbitration.

Essentially, Longoria is now set for life. No matter what happens, he can bank on receiving $17.5 million. Before he signed this deal, if he had gotten hurt during the next three years he would have made less than a million dollars. He does give up the chance to sign an outrageously lucrative contract when he's 28, but for any young player, divesting yourself of all that risk is too great an opportunity to pass up. This is a deal he should take ten times out of ten.

The Rays have assumed all of Longoria's injury and performance risk, but at a very affordable rate: less than three million dollars a year. Longoria is as "can't miss" as prospects come, but even if he does "miss," the Rays would have to make five bad deals of this nature for every success in order for them to be substantially hurt. On the other hand, if Longoria doesn't miss, they have a potential All-Star locked up for nine years at less than five million dollars a year. That's an absolute steal for them.

This is the new baseball market. I think it's awesome.

** EDIT ** The discussion going on here, in which I am a heavy participant, is worth reading, if only to see other people's reactions.

Thursday, April 17, 2008

Does Manny kill Yankee pitching?

From Pete Abraham's Yankee Blog:
Manny Being Manny is an insane 52 of 110 (.473) against the Yankees since the start of the 2006 season with 12 homers and 35 RBI in 32 games. He has 53 homers and 153 RBI against the Yankees in his career.
Later in the same post:
He’s .320/.408/.615 in 193 career games against the Yankees and .319/.397/.597 at Yankee Stadium. A .408 career OPB against one team? That is not a small sample size, either.
Two points:
  1. Believe it or not, the evidence from the first selection still constitutes a small sample. One hundred and ten at-bats is not enough to sufficiently eliminate the statistical noise present from random variations.
  2. Manny is a career .313/.409/.593 hitter. He's hit a few more singles against the Yankees, certainly nothing that would cause us to suspect that he kills them.
Manny kills everyone. It is a treat to watch him bat (and field too, but for different reasons). This is not a Yankee related phenomenon.

Saturday, April 12, 2008

Tim McCarver: Still doin' his thing

Tim McCarver, while inexplicably defending a horrible baserunning gaffe by Jason Varitek, on how baserunning mistakes are made (paraphrasing):
Most baserunning mistakes happen when a runner does not try to take an extra base, not when he gets thrown out trying to take an extra base.

Got that? Not making an out: bad. Making an out: not as bad. Somehow, I really think that this is not the case. In fact, even if it is literally true that the greater quantity of mistakes are passive, the enormous gap in the magnitude of the mistakes makes the point totally worthless. Given that you are going to make a mistake, make the mistake that doesn't cost your team an out.

Friday, April 11, 2008

Journalistic integrity

From Rob Neyer's ESPN Insider blog, comes a link to a Philadelphia Inquirer editorial about the costs and benefits of publicly financed stadiums:

Why, then, given the overwhelming academic research challenging stadium-centered economic development do political leaders (if not average citizens) still support such projects? In a just-released article in the Journal of Sport and Social Issues, my colleagues and I studied media coverage of 23 publicly financed stadium initiatives in 16 different cities, including Philadelphia. We found that the mainstream media in most of these cities is noticeably biased toward supporting publicly financed stadiums, which has a significant impact on the initiatives' success.

This bias usually takes the form of uncritically parroting stadium proponents' economic and social promises, quoting stadium supporters far more frequently than stadium opponents, overlooking the numerous objective academic studies on the topic, and failing to independently examine the multitude of failed stadium-centered promises throughout the country, especially those in oft-cited "success cities" such as Denver and Cleveland.

I'm shocked - shocked - to find that people tend to be biased towards views in which they have a significant vested interest.

Thursday, April 10, 2008

Low hanging fruit

I try to refrain from directing my readers to the various mainstream media fiskings over on FJM. While often hilarious, they don't constitute (and are not meant to constitute) serious baseball analysis and are generally heavy on profanities. This particular send up of an almost astonishingly ignorant view of baseball (and perhaps life in general) is no exception to these rules, but if you were just thinking that the one thing you could really use right now was an outrageous dismantling of an even more outrageous piece of horrible baseball writing, your wish has been granted.

As my only comment on the substance of the article itself allow me to simply say that I think that baseball is big enough for everyone who wants to like it, regardless of how one derives the the most enjoyment out of it. So for those of you who want to talk about RBIs, Ws, Ls, and AVGs, have at it! We all love the same game, we just talk about it in different languages!

Wednesday, April 9, 2008

It's early, people.

Just a reminder that in the baseball season a week and a half in April matters just as much as a week and a half in May, June, July, August, or September. It just seems like everything is more significant because because there's no context. Nothing you have seen, think you have seen, or feel has any significance whatsoever yet!

Overreaction to super small samples: fun for the whole family!