Breaking Down the Remaining SEC East possibilities

To say the SEC East has been wild is something like saying Amanda Bynes has been weird.  There have been surprising teams (Missouri), shocking losses (UGA to Vanderbilt and South Carolina to Tennessee) and all around hilarity (did you see the world's largest outdoor cocktail party?  Where the happenings on the field certainly lived up to the game's moniker?).

As far as figuring out how this could all play out, there hasn't been a whole lot of clarity.  The various options that could happen seem to be confusing enough, let alone to get an intuitive grasp on how likely any of them in particular are.  However, not that many games remain for the teams involved, so it is actually possible to get an idea of what can happen, who is in the best shape and how likely any given outcome is.

The only SEC games that remain for the teams involved are Florida at South Carolina, Missouri at Kentucky, Missouri at Ole Miss, Texas A&M at Missouri, Kentucky at Georgia, and Georgia at Auburn.

This season, Kentucky has been extremely inept, still winless in the SEC, and only two wins overall, versus Miami of Ohio and Alabama... State.  That makes things a bit clearer, as we can ring up near certain losses v Missouri (I gave Missouri a 90% chance of winning this game, it is on the road after all) and Georgia (I gave UGA a 95% chance of winning this game, anything can happen, hence the 5%, but it's extremely unlikely).

This means there are only 4 real games that would majorly swing this race remaining, I'll deal with them in turn:

1) Florida at South Carolina (75% chance South Carolina wins)

This is a rivalry game, though not on the level of Florida v Georgia.  Most anything can happen, but Florida is banged up, hilariously inept at times on offense, and has a reeling defense.  And they're going on the road to Columbia, SC, where the Gamecocks haven't lost since 2011.  I gave South Carolina a 75% chance of winning this game.  Florida has a lot of talent, but South Carolina will likely have more talent that is better coached and at home.  It's not a game South Carolina can take for granted, but it is one they should certainly win, especially coming off a bye week.

2) Georgia at Auburn (40% chance Georgia wins)

Auburn has been the surprise team out of the SEC West, with just a road loss  to LSU sullying their schedule and impressive road win over Texas A&M.  The South's oldest rivalry is going to be HUGE for both teams, as a UGA loss severely impacts their chances, and Auburn would be playing for a chance to take the SEC West with a win in the Iron Bowl.  Given the way the two teams have played, and the quality of their wins (and in Auburn's case comparative lack of losses, let alone bad losses), giving Georgia a 40% chance here might even be generous.

3) Missouri at Ole Miss (50% chance Missouri wins)

Look at Ole Miss's schedule.  They don't have a bad loss, losing to Alabama, Texas A&M by just three and on the road at Auburn by just eight.  Add in a three touchdown victory over Texas on the road and a victory against LSU and you see what is a very under the radar, very good team.  Ole Miss has been very tough in the land of Hotty Toddy, and I don't see this being anything other than a coin flip game for Missouri, and that might be generous.

4)  Texas A&M at Missouri (30% chance Missouri wins)

Missouri gets TAMU at home, but TAMU is still TAMU, and they still have perhaps the single most valuable player in college football in Johnny Football.  TAMU has had some slip ups, but they're probably still a better team than Missouri at this point.  Further, Missouri's home field advantage isn't really on par with South Carolina and Georgia, so they don't get as much of a home field advantage bump as either of those two might in this situation.

Given those games (plus the two games against Kentucky), let's look at how it all plays out:

Chances of winning the SEC East outright:

South Carolina: 19%
Missouri: 21%
Georgia: 4%

As we can see, the chance that we get an outright winner of the division is under 50%. Meaning more likely than not, we'll see some sort of tie break scenario come into play.  UGA is hurt by having more losses than Missouri already, and a tougher schedule with more games to play than South Carolina.

Two team tie scenarios:

Georgia/Missouri tie (Missouri wins by virtue of head to head win): 4%

This outcome is pretty unlikely because it not only involves UGA and Missouri ending with the same number of SEC losses, but also South Carolina not having that same number.  When you multiply all those things out, it ends up being pretty unlikely, which means that Missouri's win over UGA doesn't mean as much as it might have.

Georgia/South Carolina tie (UGA wins by virtue of head to head win): 12%

This is a somewhat unlikely outcome, but not totally unreasonable either.  It's more likely than the above scenario because while USC can only have two or three SEC losses, Missouri can have 1, 2, 3 or even 4.  Meaning it's more likely that Missouri doesn't send this back to a three way tie.

USC/Missouri tie (USC wins by virtue of head to head win): 22%

As we see, as two team tie between the Tigers and Gamecocks is among the more likely of the specific outcomes, making South Carolina's road victory over Missouri huge.  Given the games Missouri has left to play, their most likely outcome is having two SEC losses (46% according to the outcomes of the games above), combining this with the chances that UGA loses on the road at Auburn and USC beats Florida at home, and we see just how big Connor Shaw's 4th quarter come back could be.

Three way tie scenarios:

Missouri has fewer SEC East losses (Missouri wins): 13%

Because the three teams beat each other circularly, the first tie breaker in a three way tie would be who has the best record in the SEC East.  Since Missouri's two toughest games remaining are against the SEC West, not the East, this is something that could heavily play in their favor.

Pandemonium (AKA would involve all sorts of games that aren't the three listed teams, and maybe even BCS rankings): 6%

This is the case where a 3 way tie happens and the teams involved have the same SEC East records.  Because this case would involve so much that we can't really know at this point, or even really speculate about without going into crazy amounts of math, I just simply divided the likelihood of this one up by three for each of the teams.  It's a small amount anyway, and wouldn't swing any of the numbers by any dramatic amount.  But knowing the SEC East this year, the least likely, most confusing outcome could well happen.

What it all means

I finally added up all these scenarios' probability for each team (except the pandemonium scenario, which I gave each team an equal share, 2%, for the possibility of that scenario), to get an idea of how likely each team is to win the SEC East:

South Carolina: 42%
Missouri: 40%
Georgia: 18%

As we see, this basically means that Missouri and South Carolina are both about dead even, and both are about twice as likely to win as Georgia.  This makes some intuitive sense, as while Missouri has the fewest losses, they also have a difficult schedule with a lot of remaining games.  South Carolina has one more loss, but only has one more game, that they should win.  Georgia is in the unfortunate position of being behind in both losses and also not having a particularly easy schedule.

If nothing else, I hope this kind of clarified what could happen.  The particular numbers aren't as important as realizing that some of these outcomes are unlikely because so many other things have to happen.  UGA's position is lessened for this reason a lot, because they simply need a lot of events to come up their way to get in.  While Missouri controls it's own destiny, it has a lot of very difficult destiny to control.  South Carolina has maybe the best chance, but it isn't even more likely than not that they end up in Atlanta.

It'll be a fun, and perhaps heartbreaking race to watch play out.  

If you want to see the spreadsheet I used to make these calculations, you can see it here, and you can even download it and put in your own numbers for the 6 SEC games left for these 3 teams, and get the odds based on your own estimations of how likely they are to win.

Protest of June 17th game formally denied

As expected, Fredi Gonzalez's formal protest of the Atlanta Braves 6-2 loss to the Texas Rangers on June 17th has been formally denied.

The protest was based around two plays, one coming in the 5th inning and one coming in the 9th inning. Gonzalez was also said to have protested the batting order the Rangers sent out as well.

First, in the fifth inning, with Elvis Andrus on 2nd and nobody out, Josh Hamilton did not bunt. Fredi immediately protested the play from the dugout, yelling at home plate umpire Larry Vanover "Larry, you know as well as I do he has to bunt the runner over there." Making matters worse Hamilton then singled Andrus in, which enraged Fredi even more. Asked about the play after the game, Gonzalez said "well, it was clearly illegal, we know that with a man on 2nd and no outs you have to move the runner over. I know those guys are used to playing in the American League, and maybe their rules are different, but that's not how the game works here, this is a National League Park, am I right? I was about to take my normal 5th inning nap and just barely caught it, but boy was I angry."

Later in the ninth inning the Rangers held a 4 run lead to open the frame. At that point the Rangers brought in Neftali Feliz. At that juncture Fredi is said to have gone ballistic and was nearly thrown out of the game for yelling expletives at both the Rangers and umpires. Gonzalez said later "At that point they had to be just taunting the rule book, in my years of baseball, I've never seen such clear disregard for the game. I read an article on Bleacher Report saying that Feliz was their closer. That was clearly not a save situation, I know the math on that can be hard, but I asked like 5 people in the first row, just to be sure, and they all said it wasn't a save situation. And after the game the little "S" didn't show up next to Feliz's name, so I know it wasn't. Just a blatant disregard of the rules of the game."

Finally, as the Braves skipper was looking up at the scoreboard to see if the little "S" would pop up next to Feliz's name, he noticed something fishy. From watching some of the game he had clearly seen that Elvis Andrus was faster than Ian Kinsler. Yet Kinsler batted leadoff while Andrus batted in the second spot. "That was the last straw there" the manager said "I mean are they trying to deny that Andrus is the faster of those two players? They were definitely trying to pull a fast one over us, but luckily I caught it at the end of the game." Fredi continued "I know these protests don't go anywhere most of the time, I protested a game last year where the opposition batted a third baseman second, and that was denied. But this was three terrible, blatant instances of a complete disregard for the rule book. They can't just let this go unpunished."

When the letter finally arrived announcing that the Braves' petition had been denied the manager and coaches were engaged in a friendly game of "guess whose OBP is higher" where nobody had gotten the correct answer in 6 rounds, since Terry Pendleton correctly guessed that Brian McCann's OBP was higher than Tim Hudson's, though Fredi said "man, it seems like it feels like a lot closer than the numbers say." The coaches seemed unsurprised at the result of the letter, most admitting to not remembering the game at all anyway. Pitching coach Roger McDowell noting "hell, I've probably drank 36 handles of Jim Beam since then and beat up four gays, how the hell am I supposed to remember that kind of shit?"

Fredi read the letter aloud: "Attention Atlanta Braves Organization, your petition of the June 17th Game versus the Texas Rangers has been formally denied. Are you trying to troll us with this crap?"

Breaking Down Beltran, Pence, Heyward to the Minors, Upton, etc

I said yesterday that Beltran to the Braves just wasn't going to happen. Who knows how close it ever came and if there ever was a chance, but I get the distinct impression that as soon as both the Rangers and Giants were involved, the Braves were essentially out.

The analysis of the trade in the blogosphere seemed to begin with what teams could offer the Mets. However, I think an underappreciated angle was how much influence Beltran and Boras had over the process.

Beltran had indicated only really three things: 1) that he wanted to play for a contender (duh, who else would trade for him anyway?) 2) That he would prefer to play RF and 3) that he wanted to play everyday.

The problem with the Braves is that they seemingly couldn't really guarantee parts 2 and 3. In an interview with David O'Brien yesterday, Boras sounded for all the world like the Braves weren't even really a consideration. If the team wasn't giving up Minor, that wouldn't make Boras talk that way, Boras doesn't care who the Mets got, as long as it was enough to get his client moved. No, it's my opinion that once The Rangers and the Giants, both contenders who could gurantee Beltran an everyday spot in RF, got involved Beltran indicated he'd probably veto any trade to the Braves, who could only guarantee an everyday spot if Beltran would play CF at least part time. I actually doubt that once those two clubs were involved, the Braves actually had much of a chance to not offer Mike Minor. Once Boras saw that there were two clubs willing to offer a package that the Mets would accept, he essentially represented his client's wishes by indicating that the club that couldn't offer an everyday spot in RF was done. Classic game theory. Of course it's just speculation and we can never know for certain, but I think it's plausible if not probable.

Now, what of this Hunter Pence posturing? Dufus extraordinaire, Jim Bowden, of ESPN has said (along with other reports) that the Braves are heavily in on Hunter Pence. He also indicated (again, aping other reports) the Braves would option Heyward to the minors in order to open up a slot for Pence. If they are willing to do this, why wouldn't they give up Minor for Beltran and say we'll option Heyward to the minors?

It's my belief that this is either completely fabricated by bored beat writers (mostly Mark Bowman, who seemed to completely pull the original Heyward to AAA rumor out of his ass) or it's a deliberate plant by the Braves, to create leverage with the Rays and Cubs, who have CFers that the Braves likely covet in BJ Upton, Reed Johnson and Marlon Byrd.

The problem with the Braves position is that they clearly want a RH upgrade in CF, but the options are extremely slim. So perhaps they're creating an illusion of options so that they have a bit more leverage with the Rays with Upton and the Cubs. If it seems like they might go with Pence, perhaps the asking price for Upton, Johnson/Byrd goes down. However, if the Rays know that Upton is really the Braves only option, the asking price could get astronomical.

I firmly believe however that the Braves would make no move before they'd trade for Pence and send Heyward to the minors. Again, if they would be willing to trade for Pence, they would have made a harder push for Beltran. Their actions indicate that they actually want a CFer first and foremost and are very unlikely to trade for an everyday corner OFer.

I'd rate the major remaining options as:

BJ Upton: 25%
Marlon Byrd: 15%
Reed Johnson: 30%
Hunter Pence: 10%
other: 10%
No move: 10%

Of course if the Braves do trade for Pence and option Heyward, this is all hogwash. I also will consider being done with the Braves this year.

If I'm The Wrenmaster

First, main needs, in relative order of importance:

1) RH reliever(s) - Yes, more important than a bat at this point. The lineup may or may not come around, but at current usage rates, our top relievers will fall apart if they don't get help. Moylan will help when he comes back, but he's no guarantee, and we need more than that anyway.

2) RH bat - Yes, even if Uggla comes around we need another RH bat in the every day lineup.

3) CF - Schafer is not the answer, McLouth is passable but we need more offense here, ideally not sacrificing defense.

4) utility guy - Lugo getting significant playing time is a problem, though not a HUGE one.

5) RH bench bat - Because there is a chance that McCann could get injured after Ross PH's, meaning the Braves would have to forfeit all subsequent games.

First target is a RH RP. One legit guy along the caliber of Rauch or Clippard. Whichever is cheaper. Also bring up one of Teheran and Vizcaino. It's getting to the point where they should be scaling back IP's anyway, might as well be doing that in the Atlanta pen. In a sort of backwards way of thinking, if you think Teheran will be in the opening day rotation next year, you bring him up, since he wouldn't be scheduled as a super 2. If you think he won't then bring up Viz. For the playoffs I'd actually consider bringing them both up. How sick would: Teheran, Vizcaino, Moylan, O'Flaherty, Rauch, Venters, Kimbrel be? That bullpen might give up negative runs. You could easily afford to take your starters out before they face the lineup a 3rd time, every time. Venters and O'Flaherty as lefty specialists?

Trade for Upton, Marlon Byrd or Reed Johnson, in that order. Carlos Beltran is nonsense, even if he is willing to play CF, what kind of defense do you expect? Can we afford to have both him and Chipper in the lineup, depending on their bats with all that instability? You're not going to fucking demote or platoon Heyward, get the fuck over that.

Upton is my first choice, as we'd see the defense in CF actually improve, and he's got decent OBP, power and speed skills. He's not a leadoff hitter, despite his speed, but you deal with that. He's both a RH bat and a plus plus plus defender in CF. And Jordan Schafer would be gone. I'd go Minor + any of our position players besides Pastornicky or Salcedo. I haven't wrapped my head around if the Rays want Salcedo or Pastornicky straight up. Value seems right, but we need Chipper's eventual successor.

If the Cubs would act reasonably about Byrd, I'd target him, but Hendry has shown some reluctance to trade players who are under contract for next year. Is it a bluff? Who knows?

Reed Johnson would be a rental and would come cheap and would be an upgrade in CF offensively, is RH and wouldn't be a gigantic downgrade defensively.

Trade for Jonny Gomes as a pure utility guy/RH bench bat, if we can get him for next to nothing. Otherwise, do without and fucking tell Fredi to use Ross as a pinch hitter.

Finally, I'd trade Jair Jurrjens, if you can get very good value, but that's for another post.

Craig Kimbrel shouldn't pitch with a 3 run lead in the 9th, at all.

Fredi Gonzalez's usage of his bullpen gets a lot of flack on the interwebs. Things like his refusal to use his best relievers in a tie game on the road, using Proctor in anything other than a 12 run blowout, underusing some younger arms in the bullpen while overusing others (Venters and Kimbrel), the list could go on and on.

However, this post isn't so much about criticizing Fredi as it is something almost no manager would do, so I'm more pointing out a strategy I think could be effective for teams in certain situations, and especially for the Atlanta Braves of 2011-2012.

The save is a dumb stat. If you don't agree with that statement, just stop reading, move along, nothing good will come from further reading of this post. However, the save is still highly valued in baseball, for reasons that are not entirely clear. It's an important stat in: deciding how much money free agent relievers get; it actually seems to influence many managers' strategies, like the stat somehow counts in the win column; and for the purposes of this article it is really influential in player arbitration cases.

Craig Kimbrel is a great relief pitcher, the numbers he's put up this year are among, if not the, absolute best in baseball. He's also under full team control, making the major league minimum. This is awesome for the Atlanta Braves, as it frees up significant financial resources to pursue other spending avenues within the constraints of a limited budget.

However, this will not always be the case. There will come a time when he becomes arbitration eligible.

For relief pitchers, arbitration awards are dominated by saves. Great DIP stats to the wind, for whatever reason, saves dominate. Because of this, Craig Kimbrel could be in line for a pretty nice pay bump when he qualifies for arbitration status. As things look right now, he might be not only at the top of pitchers for his age group, but might be at the top of all relief pitchers at that point.

This is a nice problem to have, because it does mean that he's pitching extremely well. However, this burden on the Atlanta Braves could be lighter (or the benefit could be greater, depending on how you chose to look at it).

Moving from arbitration pay to workload for a second, another issue is that Kimbrel is being overworked. Now, some of this is not Fredi Gonzalez's fault. The Atlanta Braves have played in a lot of very close games, necessitating using your highest leverage relivers much more than average. The Atlanta Braves are also very top heavy in the bullpen, with Venters and Kimbrel being exceptional, but having questionable relievers after that, especially right handed relievers. It'd be really nice if we could figure out some way to lighten Kimbrel's workload, without taking him out of those situations where we need him the most.

For the third prong, let's talk about the three run lead in the ninth inning and the value of throwing a scoreless inning in that scenario. We'll use Win Probability Added (WPA) to show how important (or unimportant) a scoreless 9th inning with a 3 run lead is. Using yesterday's Braves-Rockies game as an example we see Craig Kimbrell's scoreless bottom of the 9th with a three run lead was worth a WPA of .053. This isn't an entirely meaningless number, but you'll notice Jonny Venters' scoreless inning was worth more, and 6 Braves positions players had a higher WPA for the game. Basically Kimbrell's number means that his scoreless ninth inning increased the braves chances of winning the game from 95% to 100%. A 5% increase in probability. Not negligible, but not earth shattering either.

Stepping away from an advanced stat like WPA, just think about it for a second using a more flawed, yet better understood statistic, ERA. A pitcher with a 27.00 ERA is obviously historically, laughably bad. Yet, that's essentially the ERA you have to have in order to blow a 3 run lead in one inning (of course barring unearned runs). So while a scoreless ninth with a three run lead is valuable, it's not all that valuable.

Yet the three run lead in the ninth inning is treated for our purposes just like an extremely high leverage situation. It takes a toll of 1 IP + warming up in the bullpen on Kimbrel's overworked arm and adds to the amount of money he will get in arbitration.

So if we're looking for a way to cut down on the load on Kimbrel's arm, and save the team some money for a possible extension (Tommy Hanson or Jason Heyward maybe?) then it'd seem like an obvious choice. Stop putting Kimbrel out there with three run leads in the ninth inning. Let someone else pitch besides Venters and Kimbrel in those situations. You'd save your team some money by distributing those saves to guys who aren't going to be arbitration guys to begin with (either because they're past arbitration years or not going to get enough saves to matter anyway) and we can give Kimbrel some much needed rest.

Remember, saves don't win games, wins win games.

The Difference Between a Statistic and a Model

Something that has long annoyed me about baseball statistics is the misuse of terms. For instance, we talk about Batting Average and Slugging Percent. Yet, batting average is a percent (the percentage of base hits in a given number of plate appearances that result in either an out, error or hit) while slugging percent is an average (the average number of bases per at bat that resulted in a hit, error or out, not counting an error as a base). But that's really just being nit picky. Who really cares if you use incorrect terms in those regards? It wouldn't really change the way that anybody looks at either statistic.

However, there is a more nefarious misunderstanding of terminology, and it actually tends to infect those who claim to have the best understanding of statistics. That is the difference between a statistic and a model.

A statistic is simply a single measure of some attribute. Batting average is a statistic. It takes the total sample (all outcomes for all plate appearances), isolates a subgroup (plate appearances resulting in a hit, error or walk) and then puts as the numerator a further subgroup (hits). Mostly, statistics don't make value judgments. You can say that the focus on a given statistic makes a value judgment, that is choosing to pay attention to batting average emphasizes that getting hits is more important than taking walks. However, the statistic itself says "hey, here's some information, use it as you will."

A model on the other hand makes value judgments. Models take phenomena and try to make predictions. For instance, OPS is a very simplistic model. Because slugging percentage and on base percentage measure entirely different things, it doesn't make statistical sense to crudely add the two. The value comes from a belief that when you add the two you get a rough model that gives you some sense of a hitter's value that you wouldn't be able to get from either of the component statistics in isolation. Yet, OPS is often referred to as a statistic, as if it was just giving raw information. Yet, it isn't. An OPS of .800 has all sorts of ways of happening. It could be a .300 OBP and a .500 slugging percent, it could be .400 and .400, etc. The model makes a value judgment in assigning equal importance to OBP and slugging pct. When you look at OPS as a statistic, you are forced to think that OBP and slugging percent have equal worth. Whereas with batting average, you can value the statistic as little or as much as you would like, you aren't forced to value anything to any particular degree.

And therein lies the problem, by calling OPS a statistic, people are mislead into believing that value judgment. By calling it a statistic you are lead to believe "hey, I'm not saying anything myself here, that's just what the numbers say." However, if you understand OPS as a model, you can think "hey, models can be inaccurate."

Most of the new "statistics" that the sabermetric crowd has been pushing lately are models, not statistics. There is nothing wrong with building a model, but you must call it a model. Calling it a statistic, while possibly simpler, is inherently misleading. WAR is a model of predicting how many wins a player adds above a replacement level player. It's not only a model, it's a model built out of other models (for instance a model for what a replacement player is worth, the relative merits of a stolen base v. getting caught stealing, etc). Yet sabermetricians tend to say "these are the stats" as opposed to "this is the prediction the model gives." Additionally, when you know you are working with a model, it's easier to gauge how reliable the model is. Models are inherently more reliable the closer they are to the most common values. For instance WAR is probably MUCH more accurate the closer it gets to players who give relatively average component statistics. WAR is probably pretty inaccurate for a guy like Albert Pujols, who usually is near the top 5% in almost every offensive category.

Furthermore, calling a model a statistic shields you from talking about the assumptions that the model makes. For instance, in calling OPS a statistic, nobody really discusses the assumption that on base percent and slugging percent are of equivalent worth. WAR assumes that a given player is just as valuable on one team as he is on another. And that's definitely not realistic. For instance a player who hits a lot of homeruns is much more valuable on a team that has a high OBP and low slugging percent than on a team with a high slugging percent, but low OBP. War assumes that a given player will perform to the exact same level in all playing environments (for instance, flyball hitters are probably much more valuable in the AL East, where all the ballparks are very hitter hitter friendly). These effects are often minor, but can be important at the margins, which is where WAR is used most often. Nobody needs WAR to tell you that Albert Pujols is more valuable than Alex Gonzalez, we use it most where the numbers are close, which is where the numbers are most likely to fall prey to these otherwise small issues.

Now, this would be all good and fine if sabermetricians just called these models stats for convenience's sake. However, it definitely seems that very few of them understand this difference very clearly.

In sabermetric circles, it lately seems as if the more complex you can make a model, the better. Sabermetricians don't really understand that the more complex you make a model, the more things you have to finely calibrate, and the more data you need to make your model accurate. A simple model may make questionable assumptions at times, but it's relatively easy to calibrate. A complicated model needs tons of data and is only really good for the centermost values.

Emmanuel Derman, perhaps the worlds most famous living financial modeler, once said "people often forget the point of models. You can't make a perfect model, because the only perfect model is reality itself, which is too complex, which is why you wanted to make a model in the first place. You have to find the right balance of simplicity, easy to understand assumptions and relevant assumptions. A model with tons of difficult to understand and complex assumptions that may be very relevant is really no better than a simple model with easy to understand assumptions that may not be as relevant at all times. Because the more complex your model gets, the less you are able to understand when it is going to be accurate and when it is going to be inaccurate." His point was, if you build up incredibly complicated models, it often gives a false sense of accuracy, because it is so often difficult to see where the assumptions might be faulty, because there are so many of them. That's basically the philosophical mistake that Long Term Capital Management made. I often feel like that's the mistake a lot of sabermetricians make. They're so obsessed with coming up with the one model (which they call a statistic) to rule them all, to give the ultimate measure of a player's value, that they forget they're building models. They forget to keep track of where they made assumptions and talk about cases in which their model may very well be highly inaccurate. The point of a model is to make relationships easier to understand, not more obscure. Yet that is often what sabermetricians do when they create new versions of WAR. They obscure the relationships they're trying to express in order to create the one model to rule them all, the one model that spits out a perfect measure of value.

So what can be done? Well, as my econometrics professor often said, "if in doubt, return to the simplest statistics possible." WAR is comprised of a lot of simple statistics that don't make value judgments. Return to those and debate their relative importance. And if in doubt return to the simplest statistics of all, things like walks/PA, singles/PA, doubles/PA, HR/PA, etc. Then talk about how those should relatively weigh and how accurate or inaccurate the various models may be, and in exceptional cases, you probably shouldn't rely on the models at all.

The attraction to WAR is that it does this for you, it makes implicit assumptions about these various values. Yet, this is really its downfall as well. People trust these assumptions when they may very well be far from trustworthy. That's what happens when you confuse a model with a statistic.

**Yes I am aware that there are different version of WAR, they all do the same thing I am talking about here, and if anything this further proves my point that WAR is a model and not a statistic.**

On Coin Flips, Hot Streaks and the MLB Playoffs

When I was in college, I once set up the following experiment for an independent study I did in psychology and economics:

A coin flipping contest with a prize of $10.

There were 128 students involved in the contest. It's pretty easy to get college kids to flip coins for free pizza and a chance at enough money for 4 beers at the local college dive bar. It was set up NCAA Bracket style such that there were 7 rounds, with each round half the pool being eliminated until we had a final champion. One person got to flip the coin and the other person got to call it. Who got to do what was determined by drawing straws. The coin had to go at least 7 feet in the air and hit the ground.

Now the tournament was completely uninteresting and actually wasn't the focus of the study. What the students said after the tournament was what I was trying to get at.

Coin flips are obviously luck, especially when the coin flips more than a couple of times and even more so when different people are flipping the coin each time. You'd have to be borderline crazy to ascribe victory in this sort of tournament to anything other than luck.

However, in talking to the winner, and even those that had made it to the last few rounds, they all described a feeling that they could predict the coin flip. The winner going so far to say that he was basically 100% confident he was going to win the final matchup. Those who had won several rounds all ascribed their winnings to something other than pure luck usually "getting a feel for how the coin would flip and how to flip if it I was the one doing the toss." The winner actually said "how could you call that luck? I picked the correct side seven straight times, what are the chances of that happening?" Well, he's right that it is extremely unlikely that a given person picks correctly seven straight times in a coin flipping contest, there is less than a 1% chance of that happening. In fact the chances are exactly 1/128.

However, the problem is that we had 128 students and a contest set up such that somebody had to win. Stepping away from an individual in the coin flipping contest's point of view, it was clear that somebody had to be lucky and it just happened to be him. But our brain doesn't like that explanation. Our brain is designed so that, as much as possible, we don't ascribe unlikely events to dumb luck. The winner earnestly believed he had some sort of skill in predicting those coin flips.

In his career Barry Bonds came to the plate 12,606 times. In his career he reached base 44.4% of the time and got a base hit 29.8% of the times he didn't walk, sacrifice or get hit by a pitch.

Imagine a time when Barry Bonds reeled off seven consecutive hits. What do you think the announcers were saying? They were probably saying something along the lines of "wow, Barry is just completely locked in right now! That baseball must seem as big as a watermelon to him!" His manager would be elated and thinking about the carnage his zoned in slugger would be doing against that day's pitcher. Not only was it Barry Bonds, but it was Barry Bonds on a torid hot streak. Now, the fact is that while seven straight hits is highly unlikely, for a player who gets a hit 29.8% of the time and had just under 10,000 ABs, you'd expect it to happen right around 7 times. So while it was unlikely for it to be happening at any given point in time, by just pure dumb luck, you'd expect it to happen seven times over his career. Yet do players, managers, fans or announcers ascribe these sorts of streaks to luck? No, they ascribe them to being hot and skill.

Now imagine we flip the script and Barry Bonds has just had back to back 0-4 days, zero for his last eight. The announced would be talking about how Barry needed to make x adjustment, that his timing was off, that he wasn't seeing the ball. His hitting coach might even suggest some sort of change. He would be cold. Except for he wouldn't. When a player doesn't get a hit 70% of the time, there's roughly a 5% chance that he will go a couple of games without getting a hit. in a full season, you'd expect that to happen several times, just by pure, dumb, bad luck. Yet, we almost always view it as the hitter is cold and may actually need to make some sort of adjustment to his swing.

To be sure, skill or lack of it has something to do with it. A player who gets a hit 35% of the time is much more likely to have a long hitting streak than a player who gets a hit just 25% of the time. But the majority of it is luck. You can think of it this way, if you do something (come to bat) enough times, highly improbable things (really long hitting streaks, really ice cold streaks) end up being likely to happen at some point.

Going back to my coin flipping contest, let's imagine a sport where the absolute best teams in the league win just 60% of their games. Now let's imagine we created a tournament where we took only those teams that already didn't win all that much more than half the time, and had them play only other similarly good teams (that won between 55% and 60% of their games) in one five game series to open and then seven game series subsequently.

Think about that for a second, we are only pitting teams that win between 55% and 60% of their games against one another in relatively short playoff series. You'd expect the best teams to have very slightly better odds, but mostly it would just be luck. Teams that are that closely matched would take several hundred games of playing one another to really sort out which team was actually better. Yet, you can bet, just like the coin flip participants, that those that went out early would be called chokers, and the champion would have put together a magical run and earned their way alongside the greats of the sport. Nobody would characterize the World Series winner (oops, did I let that slip) as just lucky, they would have ridden dominant pitching and "clutch hitting" all the way to glory. Because that's how our brains work. While it's risky to bet on any given winner, it's a virtually guaranteed bet that the winner, and especially the winner's fans, won't think the win was pure luck. When betting on chance, the only certainty is that people won't believe they're betting on chance.