All right, so we’re doing something a little bit different this week. (In addition to putting the post up on Wednesday instead of Monday.) Scoring margin is better than straight-up win-loss record as an indicator of performance–and predictor for future results–but Pythagorean win-loss is even better. So this week I busted out the calculator and figured out how the Pythagorean calculation sees the SEC standings.
Teams below are ranked by their collected Pythagorean wins (or “expected” wins), with their actual wins, difference off the Pythag (i.e. how lucky or unlucky they’ve been), and runs scored/runs allowed afterward. Enjoy:
1. South Carolina, 14.988. (Actual W’s: 16, difference +1.012, RS/RA 150/95)
2. Auburn, 13.873 (W’s: 12, -1.873, 173/124)
3. Arkansas, 13.699 (W’s: 14, +.301, 163/119)
4. Florida, 13.153 (W’s: 15, +1.847, 123/95)
5. Vanderbilt*, 11.680 (W’s: 10, -1.680, 96/76)
6. Ole Miss, 11.136 (W’s: 14, +2.864, 128/136)
7. LSU, 11.056 (W’s: 11, +.056, 155/147)
8. Alabama, 9.474 (W’s: 9, -.474, 136/150)
9. Kentucky, 8.927 (W’s: 7, -1.927, 135/157)
10. Tennessee, 7.780 (W’s: 8, +.020, 112/146)
11. Miss. St., 7.054 (W’s 5, -2.054, 133/187)
12. Georgia*, 3.834 (W’s: 3, -.834, 89/177)
*Note that Vandy and Georgia have only played 19 games, so it’s a little apples-to-oranges ranking them by straight wins. But they wouldn’t have shifted in these rankings working by win percentage, so I left it alone. Just know that Vandy’s much closer to Florida and Georgia to MSU than it might look like.
So there you go.
OBSERVATIONS
— Not a ton of differences between this ranking and the straight scoring margin efforts, but it does show precisely how unfortunate Auburn’s been this season; at nearly two wins under their Pythag, they’ve been the best team in the West by a narrow margin over the Hogs, and “should” be even atop the division rather than two games back. Only Miss. St. and Kentucky have been unluckier, mathematically speaking.
That’s not to say Auburn’s season has been some sort of tragedy; two games over the course of 21 isn’t that big a deal, there’s still three series for things to even themselves out, and missing Florida on the schedule remains a huge break. The important thing is to note that unlike the conventional preseason wisdom that would see Auburn being this close to the division lead (and ahead of LSU) as a fluke, the opposite is true. We’re 7 series and 21 games into the SEC schedule. Auburn is every bit as good as anyone in the league (save Carolina), and the fluke is that they’re not any higher ranked than they are.
— South Carolina’s on the fortunate side to have reached 16 wins already, but they’ve still been the best team in the SEC by a fairly wide margin. That game’s lead at the top of the conference standings has been earned. (Still, what’s with giving up 28 runs to the Tide last weekend? They hadn’t given up more than 15 in any other SEC series this season, and it’s not like the Tide are a collection of bashers.)
— I’ve been harping on Ole Miss for weeks, and here you go: according to the Pythag, the Rebels are not only the luckiest team in the league, they’re nearly a full game luckier than any other team is lucky or unlucky. With series against Arkansas and Auburn still to come, you have to think they’ll come back to the pack.
— It’s not news in this space, but LSU–ranked as high as No. 2 in the country just a few weeks ago–hasn’t been a victim of circumstance in slipping to fourth in the West. They’re just not that good.
— Florida hasn’t been as good as their record–only the Rebels are more fortunate–but damn, outscoring LSU by 14 runs over the course of a weekend is quite the performance, no matter how badly the Tigers have collapsed.
— Poor Vandy. Not only were they already more than a game behind their Pythag, but they lost two huge potential wins when that series with Georgia was rained out. They could easily be sitting on 13 wins or so by now.
— Broadly speaking, it appears that the league isn’t quite as unbalanced as the conference standings would suggest. Their split between the 7 haves and 5 have-nots isn’t all that inaccurate, but the width of the gap between the two is–most of the have-nots ought to be a little better, record-wise, and most of the haves a little worse.
Photo by Leffie Dailey.
Nearly 3 extra wins for miss. That’s absurd. Here’s hoping we win 7 of our last nine and put this thing to bed
I’ve read (can’t find the article) that most experts postulate that two factors contribute to why a team may outperform or underperform their Pythag record: manager and bullpen. In other words, the fact that Auburn has underperformed its Pythag by so much is a combination of a poor manager and a poor bullpen. Obviously, I don’t believe this is necessarily the case (at least re: manager), but it’s interesting to hear. Also, it’s still only 21 games, which is essentially a sample size of nil when it comes to baseball statistics.
J.D., I also think I’ve seen that floated somewhere, but I think that was an explanation for teams that consistently (i.e. over looooooong stretches of time) underperform their Pythag. I’d feel fairly confident that most statheads would call Auburn being 2 games through 21 pure, unadulterated luck. I think.
Jonesy, yeah, I’m thinking 7-2 as well. But I also think 8-1 is more likely than 6-3.
That’s true. I’m surprised to learn we’re 9-5 in 1-run games over the whole season (4-3 SEC). Seems like when we’ve won, it’s been by a ton, and when we’ve lost, it’s been a close game.
I hate to look at the negative, but I hate to see that MSU is two games below their Pythagorean expectations. I hope they don’t start to equal that out against us. Then again, maybe it just means they are not as bad as their record shows, not that they are “due.” If they had 7 wins, we would still think we should beat them in the weekend series.
Yeah, I thought about that too, TX. But it just means a sweep is a little less likely. That the series is in Auburn is a help, too.
I’d venture a guess that it has a lot to do with your bullpen. A few years back, in 2007, we had a ridiculous number of 1 and 2 run losses. In fact, we had very few losses that year at all where we lost by more than 1 or 2 runs.
In fact, here is a break down of the 25 losses we had that season:
1 runs – 14
2 runs – 3
3 runs – 3
4+ runs – 5
I’d venture a guess that your stats would say that our 2007 team that went 16-14 should’ve won the SEC that year.
The problem is, our offense wasn’t good enough to separate from teams, and our bullpen had a problem closing out close games all year.
I know Auburn’s offense is great, so I’d venture a guess that your problem is your pitching as a whole isn’t that good, and your bullpen in particular is having trouble with the close games.
I wouldn’t call it luck as much as I’d say it’s the bullpens. Our bullpen has done a pretty good job when faced with tight games. They’ve been hit hard at times, but that’s usually when we fall behind big. When they are under pressure with tight games, they’ve responded.
Rebel, thanks for stopping by. I have no doubt that with 14 one-run losses (holy crap!) Pythag would have loved that 2007 team.
If we were talking about a larger sample size, I might agree with you about the bullpen, and it’s fair to say that it’s not a strength of Auburn’s team. But without being able to tell you off the top of my head the starters vs. bullpen’s ERA, I don’t _think_ they’ve been any worse than the starters; there have been plenty of games, in fact, where our starters have gotten rocked and the bullpen’s kept us within striking distance. And over the course of 21 games, I don’t know if any factor would have enough impact for us to say for certain “that’s it.” It may in fact be the issue behind Auburn’s underachievement record-wise, but at this point I think it’s still safer to call it luck IMHO.
Fair enough, and to be honest, I’m very concerned about Auburn in the West race/hosting race. We have a big one with Arkansas this week, but based on the schedules (UT and MSU coming up for you, South Carolina and Vanderbilt for Arkansas following this weekend), I expect that the final weekend in Oxford may very well decide the SEC West, assuming we can take 2 from Arkansas this week.
I wouldn’t be surprised at all if that’s the case.
I’ll be up-front and admit I’m an Ole Miss fan, and I think that we have been one of the luckier teams in the SEC.
There’s a more advanced form of the Pythagorean Expectation in which you use ((RA+R)/GP)^0.285 which has been agreed upon by several sabermetricians to be the most “accurate.” Interestingly, I recalculated everything (leaving out Friday’s games) using that method, and Ole Miss falls to 9.846, meaning a differential of +4.154! It hardly affected the other teams.
Naturally, the small sample size and the fact that the formula was derived using MLB scoring patters makes these results more interesting than telling, but fun to consider nonetheless.
Thanks for the info, Duker! I considered using the “Pythagenpat” formula you’ve got there ,but I’m not a math major by any stretch of the imagination and figured the regular formula was complicated enough for me. Interesting that Ole Miss was the only team whose results changed that much.
Best of luck going forward, last weekend of the season excepted, naturally.