coach prestige ratings system...

[theharbinater]

Quote: Originally Posted by kevj349

The formula is good and all but a lot of it just comes down to perception. Ohio (pre-tats) view Jim as Jesus. The rest of the country admires the crap out of him. The man has some serious power behind his name. I know I'm an OSU guy but still ha

yes, and my formula has him as a top 5 coach. with 2 of the top 5 debatable in joepa and spurrier, who are both there based on accomplishment in the fairly distant past. (but both are still considered in the top5-10 coaches in ncaa history, so i'd argue their prestige is still really high).

so he's top 5 at least, and potentially top 3 if you feel joepa and sos need to drop a little.

i'm not sure i see the problem.

[schumj]

Quote: Originally Posted by Baughn3

I agree somewhat, but look at the talent that Mike Shula brought to Alabama, and compare that to what Saban has done. The coach definitely makes a difference. As for your other point, maybe you could add weight to victories in certain conferences? So maybe Big 12 victories would be weighted at 1.0, the Big Ten at 1.2, the Pac-10 at 1.4, SEC at 1.6, the ACC at 0.8, Big East 0.6, and so. Obviously the weight would be subjective to how you think the conferences should be rated from strongest to weakest, but there are some conference ratings that you could go off of

I would think weighting conference would make it way flawed. There is no way you can determine this year and and year out of a coaches career (Especially guys like Paterno) In the early 90's the SEC, BIG 8, etc were different then they were in the 2000's. Heck year by year there is a huge difference. The big east had great teams when Louisville (with Brohm and co) West Va (Pat White/Slaton) and Rutgers(Teel, Ray Rice) were all 3 some of the greatest in the nation. Three years later not so much.

[Electronic fArts]

Any rating system that doesn't have Stoops as an A+ is flat wrong. I think theharbinater's is pretty damn close to perfect.

[theharbinater]

Quote: Originally Posted by Baughn3

I agree somewhat, but look at the talent that Mike Shula brought to Alabama, and compare that to what Saban has done. The coach definitely makes a difference. As for your other point, maybe you could add weight to victories in certain conferences? So maybe Big 12 victories would be weighted at 1.0, the Big Ten at 1.2, the Pac-10 at 1.4, SEC at 1.6, the ACC at 0.8, Big East 0.6, and so. Obviously the weight would be subjective to how you think the conferences should be rated from strongest to weakest, but there are some conference ratings that you could go off of

Quote: Originally Posted by schumj

I would think weighting conference would make it way flawed. There is no way you can determine this year and and year out of a coaches career (Especially guys like Paterno) In the early 90's the SEC, BIG 8, etc were different then they were in the 2000's. Heck year by year there is a huge difference. The big east had great teams when Louisville (with Brohm and co) West Va (Pat White/Slaton) and Rutgers(Teel, Ray Rice) were all 3 some of the greatest in the nation. Three years later not so much.

you guys need to read this post: (if this isn't what you're talking abut then i apologize, and i'm not sure what you mean. if you could explain a little further then maybe we can come up with something):

Quote: Originally Posted by theharbinater

true about the more with less, thing. don't really know how to factor for that, though, so any suggestions, anyone? but for the conf difficulty, i tried to take that into account, but in a backdoor kinda way. i did 2 conf win% ratings. 1 was simple the conf win %. the other was conf win% multiplied by the conf rank (which was derived by taking the average of the BCS comp rankings for conferences last year. this is from Sagarin, Wolfe, Colley and Billingsly rating systems. i linked their websites to their names so you can check them out for yourselves.) the max conf rank is 1, and goes down by .833 (1/12, for the 12 leagues, counts indy's as a league) for each rank. here's how it worked out: SOUTHEASTERN 1.00 PAC-10 0.92 BIG 12 0.83 BIG TEN 0.75 ATLANTIC COAST 0.67 I-A INDEPENDENTS 0.58 BIG EAST 0.50 MOUNTAIN WEST 0.42 WESTERN ATHLETIC 0.33 CONFERENCE USA 0.25 MID-AMERICAN 0.17 SUN BELT 0.08 so, if 2 coaches had a .65% conf win%, one in SEC one in MWC, then they would each have 2 conf win% ratings. they would be calculated as: conf: rate1, rate2 (rate 2 factors in conf rank) = Total SEC: .65, .65 (.65*1) = 1.30 MWC: .65, .273 (.65*.42) = .923 so the sec coach would get a ratings boost based on having the same win% in a tougher conf. likewise, conf champs factors in conf rank. make sense? if you think it should be different or have a better idea, then let me know.

also, fwiw, i though about going back for the bcs guru's ranking for the last 5 years or so, but didn't really want to spend the time on that. i can do it, though, if people think it needs to be done.

[volstopfan14]

Just wondering, are you going to put together coordinator rankings?

[theharbinater]

Quote: Originally Posted by volstopfan14

Just wondering, are you going to put together coordinator rankings?

as mentioned in the op, i'm not 100% sure how to do that.

in the op i suggested just taking the hc rating and subtracting 1 to 1.5 letter grades (i.e. and A+ hc would have B or B+ oc/dc). but once you got down to the D+ rated coaches it wouldn't work much.

and, of course, for the higher regarded oc/dc, maybe bump them up another +/- rating. like malzahn at auburn. chizick is rated a B+ by the system (which is fair, imo). but malzahn is certainly not a C or C- and probably better than C+. so i'd probably bump him to B-.

same with kirby smart at bama (though his rating based from his hc (saban) would be B or B+, so i probably wouldn't change that one).

likewise, for coaches you feel are not very good, drop them abother +/_ point.

the key here is to rate them as if they had just taken over as a hc, NOT how they are as an oc/dc. cause, while malzahn and smart are 2 of the hottest oc/dc in real life and probably be rated A or A+ as oc/dc, neither would be in the top 25 hc and neither would be rated above B or B+ until they proved it.

and since there is no drop in the game for going from oc/dc to a hc, you have to account for it up front. (note, once in dynasty, since we can't edit, a couple years down the road this will not be as effective. but initially, it is the best way to go forward, imo)

another thought, and this would take a lot of work so i'm not sure i'm up for it, but it would be to take the points for (for oc) and points against (for dc) and do a rating system like above for hc's.

biggest problem with that is that oc/dc are always associated with those hc. in fact, most have only been with their current hc for 2-3 years. and i can only get hc #'s for points for/against. so using those #'s would be not entirely accurate. not accurate at all, really.

so that one is not likely to happen.

got any suggestions?

[volstopfan14]

Quote: Originally Posted by theharbinater

as mentioned in the op, i'm not 100% sure how to do that. in the op i suggested just taking the hc rating and subtracting 1 to 1.5 letter grades (i.e. and A+ hc would have B or B+ oc/dc). but once you got down to the D+ rated coaches it wouldn't work much. and, of course, for the higher regarded oc/dc, maybe bump them up another +/- rating. like malzahn at auburn. chizick is rated a B+ by the system (which is fair, imo). but malzahn is certainly not a C or C- and probably better than C+. so i'd probably bump him to B-. same with kirby smart at bama (though his rating based from his hc (saban) would be B or B+, so i probably wouldn't change that one). likewise, for coaches you feel are not very good, drop them abother +/_ point. another thought, and this would take a lot of work so i'm not sure i'm up for it, but it would be to take the points for (for oc) and points against (for dc) and do a rating system like above for hc's. biggest problem with that is that oc/dc are always associated with those hc. in fact, most have only been with their current hc for 2-3 years. and i can only get hc #'s for points for/against. so using those #'s would be not entirely accurate. not accurate at all, really. so that one is not likely to happen. got any suggestions?

I can't think of any way you could do OC/DC ratings with a formula. The only way I could see you doing it is rating each coach individually based on opinion and then taking away one point in grade. So like if you thought Malzhan should be an A+ then you would rate him B+. So no OC/DC would be rated higher than B+ and in my opinion, it would make HC hirings much more realistic. The problem is there are a lot of coordinators (especially at small schools) that most people (including myself) know little to nothing about. Also, since you are using a formula for your HC ratings, doing OC/DC ratings based on opinion would kind of kill the purpose of the HC formula.

[theharbinater]

Quote: Originally Posted by volstopfan14

I can't think of any way you could do OC/DC ratings with a formula. The only way I could see you doing it is rating each coach individually based on opinion and then taking away one point in grade. So like if you thought Malzhan should be an A+ then you would rate him B+. So no OC/DC would be rated higher than B+ and in my opinion, it would make HC hirings much more realistic. The problem is there are a lot of coordinators (especially at small schools) that most people (including myself) know little to nothing about. Also, since you are using a formula for your HC ratings, doing OC/DC ratings based on opinion would kind of kill the purpose of the HC formula.

yep, which is why i said base it on the hc ratings.

if a guy is the oc for a smaller school, it more likely he's not as good as the bigger school oc/dc. otherwise, the big schools would fire their guy and hire the smaller guys.

with the few exceptions likely being for boise, utah and tcu and the like. and those all have hc's that are fairly highly regarded as well, which show up in the rating system with boise and utah hc's being A- and tcu being B+. (if i were doing that by my opinion of them, it's probably be the exact opposite, with tcu as the A- and the other 2 as B+, but that's why i let the #'s do the talking, because i'm biased.) which means that their oc/dc are going to be rated fairly highly as well.