Sports Illustrated power rankings

[howboutthoseos]

I will never understand why we have so many threads about these power rankings. This is baseball, not the BCS. There's only one ranking that matters, and that's the standings.

Something to fill time until the game starts. That's all.

[DrungoHazewood]

I don't know about that. It seems to me that the Orioles play a bunch of close games and then get shillacked whenever one of their "bad" starters takes the hill (and against Texas). To me, that is indicative of the team needing a few better starting pitchers but not necessarily about how good the talent of the overall team is.

I don't think that's true. They're 22-28 when Arrieta, Matusz and Hunter start. That seems to be pretty well in line with the run differential in their starts, if not a little better.

[Mondo Trasho]

Next to nothing if all you care about is the current standings. Run differential is far more predictive than current standings, so if you're looking for what's likely to happen over the next couple months power rankings based on run differential are much better than current wins and losses. I know that's not kind to the Orioles, but that's beside the point.

I guess if we get to the end of the season and we have 85 wins even though the calculator and advanced calculus says we shouldn't, they'll end up taking some of those wins away.

[DrungoHazewood]

I guess if we get to the end of the season and we have 85 wins even though the calculator and advanced calculus says we shouldn't, they'll end up taking some of those wins away.

Stop it. If the Orioles win 85 games with a -50 run differential they'll be a darned lucky team, and going into 2013 they better not pull a 2007 Mariners and think they're a true talent 85 win team and they just need a piece or two to put them over the top.

[Slappy]

I guess if we get to the end of the season and we have 85 wins even though the calculator and advanced calculus says we shouldn't, they'll end up taking some of those wins away.

He's just stating the research. I mean, if you look at it objectively, there are good arguments that we have overplayed our talent. Our one run record is 20-6, which is ridiculously good. We are 10-2 in extra innings. On the other hand, we are 13-19 in blowouts (5+ runs). We've been beaten by 10+ runs numerous times (19-7, 13-1, 14-3) but our largest margin of victory is 9 (10-1 vs. Oakland).

It's "more likely" that we will play down to the run differential, but *not* a guarantee. Perhaps there are other factors at play. Perhaps Buck is unusually good at managing around his weak players while maximizing his strong ones. Perhaps a much better-than-average bullpen contributes to better 1-run and extra-inning records. Perhaps fate is smiling on us after 15 years. Who knows?

But consider this. We went 45-40 in the first half with a -36 run differential. In the second half we are 10-10 with a -24. And also, our August/September record under Showalter is 61-44.

[Frobby]

Next to nothing if all you care about is the current standings. Run differential is far more predictive than current standings, so if you're looking for what's likely to happen over the next couple months power rankings based on run differential are much better than current wins and losses. I know that's not kind to the Orioles, but that's beside the point.

What's your evidence? I've never heard of run differential being used as a predictive tool within a season. If you told me that the team would continue to run a negative run differential for the rest of the season, at the same rate as currently, then I'd agree it is unlikely the team will have a .500+ record the rest of the way. The Orioles did outscore their opponents in April, and were just -4 runs in May, so I don't necessarily take the June/July differential as predictive of August/September. The offense looks a lot better with Markakis back than it did while he was out. (That said, the run differential is -24 in that period.)

At the end of the day, pitching will decide how we do. If we allow 5.5 runs a game as we have for the last 3 weeks, forget the playoffs. But as we learned when Buck arrived here 2 years ago, pitching can take abrupt turns sometimes.

[Mondo Trasho]

Stop it. If the Orioles win 85 games with a -50 run differential they'll be a darned lucky team, and going into 2013 they better not pull a 2007 Mariners and think they're a true talent 85 win team and they just need a piece or two to put them over the top.

Agreed. I've been abundantly clear about this team from the get. We're not "objectively" talented or good but winning is possible if we catch lightning in a bottle as it were. i.e. we get lucky. So far we have.

He's just stating the research. I mean, if you look at it objectively, there are good arguments that we have overplayed our talent. Our one run record is 20-6, which is ridiculously good. We are 10-2 in extra innings. On the other hand, we are 13-19 in blowouts (5+ runs). We've been beaten by 10+ runs numerous times (19-7, 13-1, 14-3) but our largest margin of victory is 9 (10-1 vs. Oakland).

It's "more likely" that we will play down to the run differential, but *not* a guarantee. Perhaps there are other factors at play. Perhaps Buck is unusually good at managing around his weak players while maximizing his strong ones. Perhaps a much better-than-average bullpen contributes to better 1-run and extra-inning records. Perhaps fate is smiling on us after 15 years. Who knows?

But consider this. We went 45-40 in the first half with a -36 run differential. In the second half we are 10-10 with a -24. And also, our August/September record under Showalter is 61-44.

Right, but at some point there's a) luck and b) chaos that dictates how well a team does. Run differential means little if you're getting pounded by teams that outclass you but you play a bunch of close games against teams as talented as you are (or teams whose injuries have made them as talented as you are). I think the eye test is a lot more predictive than run differential or other formulas.

[DrungoHazewood]

What's your evidence? I've never heard of run differential being used as a predictive tool within a season. If you told me that the team would continue to run a negative run differential for the rest of the season, at the same rate as currently, then I'd agree it is unlikely the team will have a .500+ record the rest of the way. The Orioles did outscore their opponents in April, and were just -4 runs in May, so I don't necessarily take the June/July differential as predictive of August/September. The offense looks a lot better with Markakis back than it did while he was out. (That said, the run differential is -24 in that period.)

At the end of the day, pitching will decide how we do. If we allow 5.5 runs a game as we have for the last 3 weeks, forget the playoffs. But as we learned when Buck arrived here 2 years ago, pitching can take abrupt turns sometimes.

I just took it as a matter of common sense that if you took two .500 teams at the All Star break, one with a -50 run differential, one with a +50, that the +50 was vastly more likely to play better the 2nd half. I guess I could dig up some research, but it looks intuitively obvious to me.

[jets4ever]

I guess if we get to the end of the season and we have 85 wins even though the calculator and advanced calculus says we shouldn't, they'll end up taking some of those wins away.

Who cares what anybody writes or says about us. Heck, they've even got us behind Toronto.

So, that should show you how much those rankings mean.

[Frobby]

I just took it as a matter of common sense that if you took two .500 teams at the All Star break, one with a -50 run differential, one with a +50, that the +50 was vastly more likely to play better the 2nd half. I guess I could dig up some research, but it looks intuitively obvious to me.

I agree with you, but was just wondering if there has been research done involving in-season data. I'm relatively sure there has been some research showing that a team that significantly outperforms its pythag over a season has only a "normal" chance of repeating that feat the next season.

I guess the question is, if the Orioles are Cinderella and the stagecoach is going to turn into a pumpkin at midnight, is midnight now, or next April?

[Frobby]

Baseball Prospectus currently projects our mean likely final record at 79-83, which is 9th best in the AL, 18th best in MLB. So, 23rd does seem harsh in any event. http://www.baseballprospectus.com/odds/

[Thedan0909]

I just took it as a matter of common sense that if you took two .500 teams at the All Star break, one with a -50 run differential, one with a +50, that the +50 was vastly more likely to play better the 2nd half. I guess I could dig up some research, but it looks intuitively obvious to me.

I'm pretty sure, statistically speaking, historical run differential is directly predictive of future run differential. Making the leap to predicting future wins/losses implies several assumptions about what constitutes a win. Obviously, run differential is part of it, but there are several other important factors.

So, I tend to agree with Drungo to some extent. We have been somewhat lucky this year to have as many wins as we do because we leave such little margin for error in our wins. On the other hand, I don't think our true performance is completely summed up by our Pythagorean W/L. There are strategic decisions, injuries, and other coincidences/incidentals that separate run differential from actual wins and losses.

[SilentJames]

He's just stating the research. I mean, if you look at it objectively, there are good arguments that we have overplayed our talent. Our one run record is 20-6, which is ridiculously good. We are 10-2 in extra innings. On the other hand, we are 13-19 in blowouts (5+ runs). We've been beaten by 10+ runs numerous times (19-7, 13-1, 14-3) but our largest margin of victory is 9 (10-1 vs. Oakland).

It's "more likely" that we will play down to the run differential, but *not* a guarantee. Perhaps there are other factors at play. Perhaps Buck is unusually good at managing around his weak players while maximizing his strong ones. Perhaps a much better-than-average bullpen contributes to better 1-run and extra-inning records. Perhaps fate is smiling on us after 15 years. Who knows?

But consider this. We went 45-40 in the first half with a -36 run differential. In the second half we are 10-10 with a -24. And also, our August/September record under Showalter is 61-44.

Ding ding ding ding, here it is.

The Orioles have been involved in baseball weirdness since the end of last year.

The Orioles are 42-31 in games that are decided by 4 runs or fewer. It would be interesting to see what the run diff is in non-blowouts.

edit, jsut did it and the Orioles, in non-blowouts, have a run diff of -6 outperforming their pythag by only three games in non-blowouts.

[Lucky Jim]

Ding ding ding ding, here it is.

The Orioles have been involved in baseball weirdness since the end of last year.

The Orioles are 42-31 in games that are decided by 4 runs or fewer. It would be interesting to see what the run diff is in non-blowouts.

Nearly all bad teams will be at the bad end of blow-out run-differential.

[SilentJames]

Nearly all bad teams will be at the bad end of blow-out run-differential.

Yes but they ususally also are on the bad end of the NON-blowout games. The Orioles have been very good in non-blowout games.