FB: Old Dominion Athletic Conference

[Bombers798891]

The absolute best DI college football modelling systems barely break the 78% barrier with their in-season predictions (http://www.thepredictiontracker.com/ncaaresults.php?orderby=wpct%20desc&type=1&year=15), and none can consistenly beat the spread by over 55%.  I feel like 75% for a preseason prediction is pretty impressive. Yes, DIII is different because there are generally fewer tossup games, but we also have a much larger division with less interconference play, making comparisons more difficult.

As a point of reference, last year in the ODAC pick-'em challenge, you got 153 points over 11 weeks. With 12 games per week, 28 ODAC games, 24 ODAC non-con games, that leaves 80 out-of-conference games. At 3 points per ODAC game, 2 points per non-con game, and 1 point for the rest, that's 212 points. 153 / 212 = 72%, which is pretty good considering most of the games you're picking are going to be competitive, but I think you may have a skewed perception about the nature of predictions.

I don't know how to do this kind of analysis, but I'd love to know what the following system would generate for D-III. Maybe someone smarter than me could do it by using a conference or two

1. If the team went 8-2 or better last season, pick it to win every week*
2. If the team went 2-8 or worse last season, pick it to lose every week*
3. For every other game, pick the home team

*If both teams were better than 8-2 or worse than 2-8, see rule 3.

[jknezek]

I'd say you'd be no worse than 70%... Just a guess though.

[HansenRatings]

I'd say you'd be no worse than 70%... Just a guess though.

Agree to disagree again? It's 66.3% for 2015. Change the cutoff to 7-3/3-7, and it's 67.7%. Assume the team with the better 2014 record wins, 71.7%. If I apply this to just games featuring an ODAC team, it's 66.7%. My original model, using in-season ratings (and not preseason ratings), correctly predicted 78.6% of ODAC games in 2015, and 79.7% of all other games.

It's not that hard to pick 70% of games correctly, but the nuance in football is that the better team doesn't always win. The theoretical limit to the number of games a mathematical rating system can pick correctly is nowhere near 100%.

[jknezek]

I'd say you'd be no worse than 70%... Just a guess though.

Agree to disagree again? It's 66.3% for 2015. Change the cutoff to 7-3/3-7, and it's 67.7%. Assume the team with the better 2014 record wins, 71.7%. If I apply this to just games featuring an ODAC team, it's 66.7%. My original model, using in-season ratings (and not preseason ratings), correctly predicted 78.6% of ODAC games in 2015, and 79.7% of all other games.

It's not that hard to pick 70% of games correctly, but the nuance in football is that the better team doesn't always win. The theoretical limit to the number of games a mathematical rating system can pick correctly is nowhere near 100%.

I don't disagree with facts. In this case I was wrong.

[HSCTiger fan]

I would only expect to see statistical analysis being debated on a D3 board.  On a D1 board statistically speaking I'd say less than one half of one percent would care. On a D3 board I'd expect that number to rise to three quarters of one percent with half of those who care having played at MIT.

[Ralph Turner]

The absolute best DI college football modelling systems barely break the 78% barrier with their in-season predictions (http://www.thepredictiontracker.com/ncaaresults.php?orderby=wpct%20desc&type=1&year=15), and none can consistenly beat the spread by over 55%.  I feel like 75% for a preseason prediction is pretty impressive. Yes, DIII is different because there are generally fewer tossup games, but we also have a much larger division with less interconference play, making comparisons more difficult.

As a point of reference, last year in the ODAC pick-'em challenge, you got 153 points over 11 weeks. With 12 games per week, 28 ODAC games, 24 ODAC non-con games, that leaves 80 out-of-conference games. At 3 points per ODAC game, 2 points per non-con game, and 1 point for the rest, that's 212 points. 153 / 212 = 72%, which is pretty good considering most of the games you're picking are going to be competitive, but I think you may have a skewed perception about the nature of predictions.

I don't know how to do this kind of analysis, but I'd love to know what the following system would generate for D-III. Maybe someone smarter than me could do it by using a conference or two

1. If the team went 8-2 or better last season, pick it to win every week*
2. If the team went 2-8 or worse last season, pick it to lose every week*
3. For every other game, pick the home team

*If both teams were better than 8-2 or worse than 2-8, see rule 3.

Bombers, I invite you to join the ASC Pick'em contest using that formula.

We can test it out on a typical conference.

Thanks

[hasanova]

Just drove by Guilford to take a look at the new turf field. ... Big maroon G at the 50, Guilford in the north endzone and Quakers in the south ... looks good.

[Swish3]

I know a lot of schools have gotten turf, which looks nice, but there's something about a natural grass field that I love.  Personally, I hope CNU doesn't ever resort to turf...they have the means and ability to maintain a very nice grass field.  I'm not trying to be disrespectful, hasa, as I'm sure Guilford's turf is beautiful...just figured I'd throw in my two cents.

[jknezek]

No real surprise at the top of the ODAC coaches' poll with W&L and HSC going 1-2. Guilford comes in at 3 and then E&H and Bwater.  Not a lot of respect for RMC despite getting better all last season.

[HSCTiger fan]

I know a lot of schools have gotten turf, which looks nice, but there's something about a natural grass field that I love.  Personally, I hope CNU doesn't ever resort to turf...they have the means and ability to maintain a very nice grass field.  I'm not trying to be disrespectful, hasa, as I'm sure Guilford's turf is beautiful...just figured I'd throw in my two cents.

Every ODAC team except HSC plays on turf. Bridgewater was latest team to switch (2014 I think).

ODAC preseason poll is no major surprise. IMO any team 1-5 can win ODAC. 

[jknezek]

Football just isn't a game that needs to be played on natural grass. I like football on grass as much as anyone, but the game doesn't change on turf. Not like soccer, which needs grass, or field hockey which needs turf. Football is kind of indifferent as to the surface in my mind.

[Bombers798891]

Football just isn't a game that needs to be played on natural grass. I like football on grass as much as anyone, but the game doesn't change on turf. Not like soccer, which needs grass, or field hockey which needs turf. Football is kind of indifferent as to the surface in my mind.

I mean, need is a strong word.

But I don't believe that the game doesn't change on turf, at least a bit. I've seen some really good triple option teams eaten alive by Butterfield stadium grass.

[hasanova]

I know a lot of schools have gotten turf, which looks nice, but there's something about a natural grass field that I love.  Personally, I hope CNU doesn't ever resort to turf...they have the means and ability to maintain a very nice grass field.  I'm not trying to be disrespectful, hasa, as I'm sure Guilford's turf is beautiful...just figured I'd throw in my two cents.

None taken.  Guilford's actually had turf for almost 10 years, this is a replacement with better graphics.  With five sports (and their practices) plus other events it was hard to keep the grass in shape.

[jknezek]

Football just isn't a game that needs to be played on natural grass. I like football on grass as much as anyone, but the game doesn't change on turf. Not like soccer, which needs grass, or field hockey which needs turf. Football is kind of indifferent as to the surface in my mind.

I mean, need is a strong word.

But I don't believe that the game doesn't change on turf, at least a bit. I've seen some really good triple option teams eaten alive by Butterfield stadium grass.

Yeah that's true. W&L's first playoff team ended up facing Wilkes in what could charitably be called a swamp instead of a field. It did not go well. But that was special circumstances caused by a flood. In a well-maintained field, W&L hasn't suffered too much by playing on natural grass. Of course it doesn't happen often anymore. What are they doing at Butterfield before the option teams come in? "Forget" to cut it, or leaving the sprinklers on overnight?

[Bombers798891]

Football just isn't a game that needs to be played on natural grass. I like football on grass as much as anyone, but the game doesn't change on turf. Not like soccer, which needs grass, or field hockey which needs turf. Football is kind of indifferent as to the surface in my mind.

I mean, need is a strong word.

But I don't believe that the game doesn't change on turf, at least a bit. I've seen some really good triple option teams eaten alive by Butterfield stadium grass.

Yeah that's true. W&L's first playoff team ended up facing Wilkes in what could charitably be called a swamp instead of a field. It did not go well. But that was special circumstances caused by a flood. In a well-maintained field, W&L hasn't suffered too much by playing on natural grass. Of course it doesn't happen often anymore. What are they doing at Butterfield before the option teams come in? "Forget" to cut it, or leaving the sprinklers on overnight?

I honestly have no idea. But we're also Upstate NY, right off of a lake, and these games are usually in october. Games are going to get played in lousy weather and, on a grass field, with less than ideal field conditions. Ithaca played Salisbury at home twice, once in lousy weather and once on a perfect day, and Salisbury's running game did squat both times