Sagarin says

[Gully]

The model has margins of victory to work with, but I wonder if it actually ends up understating a team like NDSU and overstating a team like EIU?

Here is what I mean. EIU seems to keep trying to score quickly even when far ahead. This contributes to their margin of victory. NDSU seems to try to run the clock out when they get ahead. This keeps the margin of victory smaller than it could have been. Taken together, this would seem to understate NDSU's ability relative to EIU's.

In other words, the model doesn't know that NDSU goes into "run out the clock" mode.

Doesn this take make any sense to anyone or am I crazy?

[westnodak93bison]

Sagarin says Bison by 33?

Sent from my DROID RAZR using Tapatalk

[bisonaudit]

The model has margins of victory to work with, but I wonder if it actually ends up understating a team like NDSU and overstating a team like EIU?

Here is what I mean. EIU seems to keep trying to score quickly even when far ahead. This contributes to their margin of victory. NDSU seems to try to run the clock out when they get ahead. This keeps the margin of victory smaller than it could have been. Taken together, this would seem to understate NDSU's ability relative to EIU's.

In other words, the model doesn't know that NDSU goes into "run out the clock" mode.

Doesn this take make any sense to anyone or am I crazy?

You're not crazy at all. You can build these things different ways. You could set it up to minimize the squares of the variances which would tend to reward blowouts more than some other alternatives, or minimize the absolute value of the variances which would reign that in, you could constrain the maximum margin that the model will accept, etc. We don't know exactly how Sagarin's model works so it's hard to say.

[RadioBison]

The model has margins of victory to work with, but I wonder if it actually ends up understating a team like NDSU and overstating a team like EIU?

This seems to make sense intuitively, but I don't know if it holds true. I can't speak for any of Sagarin's systems, but I have my own system based mainly on the ideas I picked up in this thread. One thing I do is limit the margin of victory. If I limit the cap at 45 points, the Bison are 1.4 points ahead of EIU, but if I limit it to 10, the margin is only 0.3 points. My system seems to work much better if the cap is bigger.

[bisonaudit]

This seems to make sense intuitively, but I don't know if it holds true. I can't speak for any of Sagarin's systems, but I have my own system based mainly on the ideas I picked up in this thread. One thing I do is limit the margin of victory. If I limit the cap at 45 points, the Bison are 1.4 points ahead of EIU, but if I limit it to 10, the margin is only 0.3 points. My system seems to work much better if the cap is bigger.

Not sure what EIU's margins have been. I know they score a bunch. But we put the wood to a couple of teams too, 51 and 46 (and another 46 on Ferris).

[stevdock]

Speaking of Ferris State, since they are not included in the ranking system how do they factor in to our total??

[bisonaudit]

Speaking of Ferris State, since they are not included in the ranking system how do they factor in to our total??

I'm fairly certain they don't factor in at all.

[stevdock]

I'm fairly certain they don't factor in at all.

That's what I thought too. So for this ranking only, would we have been better off by playing a good D1 team or a crappy D1 team or does it just depend on winning by more than the expected outcome?

[bisonaudit]

that's what i thought too. So for this ranking only, would we have been better off by playing a good d1 team or a crappy d1 team or does it just depend on winning by more than the expected outcome?

c. .

[thebootfitter]

Some people showed interest in these stats, and I don't mind posting them since I am doing the analysis anyway. I plan to continue this analysis through the end of the season and see if any reasonable conclusions can be drawn then.

Overall, the average predicted vs actual point differential is even smaller this week than last (only 0.2 points vs 0.3), and the variance is down a bit with a standard deviation of 14. Sagarin correctly predicted the winner 81% of the time.

Out of 106 games between DI opponents:
--> Average difference between Sagarin's predicted point differential and the actual point differential is only 0.2 points (0.3 points last week), with a standard deviation of 14 points (16 last week).
--> (81.1%) 86 games - Sagarin correctly predicted the winner (75.9% last week)
--> (11.3%) 17 games were within 2 points of Sagarin's predictions in either direction (15.7% last week)
--> (17.9%) 13 games were between 2 & 5 points off Sagarin's predictions (12.0% last week)
--> (30.2%) 23 games were between 5 & 10 points off Sagarin's predictions (21.3% last week)
--> (12.3%) 17 games were between 10 & 15 points off Sagarin's predictions (15.7% last week)
--> (28.3%) 38 games were greater than 15 points off Sagarin's predictions (35.2% last week)