CFB51 College Football Fan Community
The Power Five => Big Ten => Topic started by: Cincydawg on June 25, 2019, 02:49:22 PM
-
Home Field Advantage.
Let's stick only to competitive games, not OSU hosting Akron. And of course it may be different at NW relative to Iowa, or Penn State. Imagine some theoretical world where Penn State - having a good year - plays Ohio State 100 times, 50 at each site. What would the average scores be in you estimation?
I dimly recall The Bobs had done an analysis that showed the difference in conference games only. It was smallest in the SEC and larger in the B1G, something like 3 points versus 6 points.
So, IF it's 6 points, OSU would average a score of say 35-29 at home and Penn State would average the reverse score. I think.
(PSU at night looks like one of the toughest venues in sports, to me.)
-
Home Field Advantage.
Let's stick only to competitive games, not OSU hosting Akron. And of course it may be different at NW relative to Iowa, or Penn State. Imagine some theoretical world where Penn State - having a good year - plays Ohio State 100 times, 50 at each site. What would the average scores be in you estimation?
I dimly recall The Bobs had done an analysis that showed the difference in conference games only. It was smallest in the SEC and larger in the B1G, something like 3 points versus 6 points.
So, IF it's 6 points, OSU would average a score of say 35-29 at home and Penn State would average the reverse score. I think.
(PSU at night looks like one of the toughest venues in sports, to me.)
There are two different questions here, IMHO:
First is a question that gamblers care a LOT about but that is otherwise irrelevant. Specifically, how many points is HFA worth regardless of who actually wins the game. IMHO, the answer to this question is about what the gamblers say, 3-5 points (as compared to a neutral site so 6-10 points swing from one home field to the other).
The second question that gamblers don't really care about but that the rest of us do is how many games will HFA help you win or help your opponent beat you?
Years ago we did an multi-year analysis of home vs road records for Big Ten teams. What we found then was that all teams had better records at home but the difference tended to be greatest for teams that typically finish near the middle of the pack. That makes sense if you think it through.
If we suppose (per ELA's thread) that Ohio State will be the best team in the conference, Michigan State #7, Minnesota #8, and Rutgers the worst then:
HFA will likely not matter in many games for the Buckeyes and Scarlet Knights because they are either good enough (tOSU) or bad enough (RU) that the location of most games is irrelevant. Ie, HFA likely only matters for the Buckeyes in games against the #2 (M) and #3(PSU) teams. Similarly, HFA likely only matters for the Scarlet Knights in games against the #13 (IL) and #12 (UMD) teams.
It is much different for the middle teams. For the Spartans and Gophers HFA is likely to matter in games against each other as well as #6 (NU), #5 (UW), #9 (PU), and #10 (UNL).
Thus, teams in the middle could have as many as five games in which HFA might be enough to tip the balance while teams near the top or bottom probably only have two or three.
-
The other thing about HFA that I always push but that most people don't realize is that it is different for different teams. What I mean is that since HFA only matters (in terms of winning vs losing) in games against teams that are relatively equal, then the best schedule is one in which you play your equals at home.
Ie, if your team is a league title contender then you want to play the other title contenders at home and the cellar-dwellers on the road. However, if your team is a cellar-dweller you want to play the other cellar-dwellers at home on the title contenders on the road. Similarly, if your team is middle-of-the-pack you want to play the other middle-of-the-pack teams at home and the title contenders and cellar-dwellers on the road.
Nebraska's and Purdue's 2019 league schedules illustrate this point nicely. Using ELA's rankings, the Boilermakers play their best four opponents on the road and their five easiest opponents at home. Conversely, the Cornhuskers play their best four opponents at home and four of their five easiest opponents on the road. If both teams are title contenders, I would pick Nebraska over Purdue without hesitation because Nebraska's schedule is MUCH better for a contender. However, if both teams struggle to attain bowl eligibility, I would pick Purdue to get to a bowl because Purdue's schedule is MUCH better for a cellar-dweller.
-
There are two different questions here, IMHO:
First is a question that gamblers care a LOT about but that is otherwise irrelevant. Specifically, how many points is HFA worth regardless of who actually wins the game. IMHO, the answer to this question is about what the gamblers say, 3-5 points (as compared to a neutral site so 6-10 points swing from one home field to the other).
The second question that gamblers don't really care about but that the rest of us do is how many games will HFA help you win or help your opponent beat you?
Years ago we did an multi-year analysis of home vs road records for Big Ten teams. What we found then was that all teams had better records at home but the difference tended to be greatest for teams that typically finish near the middle of the pack. That makes sense if you think it through.
If we suppose (per ELA's thread) that Ohio State will be the best team in the conference, Michigan State #7, Minnesota #8, and Rutgers the worst then:
HFA will likely not matter in many games for the Buckeyes and Scarlet Knights because they are either good enough (tOSU) or bad enough (RU) that the location of most games is irrelevant. Ie, HFA likely only matters for the Buckeyes in games against the #2 (M) and #3(PSU) teams. Similarly, HFA likely only matters for the Scarlet Knights in games against the #13 (IL) and #12 (UMD) teams.
It is much different for the middle teams. For the Spartans and Gophers HFA is likely to matter in games against each other as well as #6 (NU), #5 (UW), #9 (PU), and #10 (UNL).
Thus, teams in the middle could have as many as five games in which HFA might be enough to tip the balance while teams near the top or bottom probably only have two or three.
well said and I agree 100%
-
Ooow, I like your thought about "teams near the middle" and the impact on their record.
-
Ooow, I like your thought about "teams near the middle" and the impact on their record.
I think there is another reason that HFA matters more for teams in the middle and that is the MAJOR upsets. Once in a while a shocking upset happens and most of those involve a much better team losing on the road to an obviously inferior opponent. Ohio State's losses the last two years are a good example of this:
- In 2017 an Ohio State team that went 8-1 in the B1G and 12-2 overall lost on the road to an Iowa team that finished 4-5 in the B1G and 8-5 overall.
- In 2018 an Ohio State team that went 8-1 in the B1G and 13-1 overall lost on the road to a Purdue team that finished 5-4 in the B1G and 6-7 overall.
Teams in the middle not only have more "relative equals" where HFA is likely to matter, they are also close enough to the contenders that they *MIGHT* knock one off at home and close enough to the cellar-dwellers that they *MIGHT* lose to one on the road.
-
As an avid CFL watcher, they do a lot of H-A series in back to back weeks, and seems like it is generally about a 30 point swing. As such I find that HFA is rather underrated.
In CFB we are only able to compare from one season to the next, where they are completely different teams. In the NFL they play the divisional opponents H-A, but rarely if ever in back to back weeks. Teams ebb and flow throughout the year, so it throws off the data a little.
But in the CFL, it is the same two teams playing a game against each other in back to back weeks, and you hardly ever see one team sweep the other. It is almost always a split with the home teams winning, and often times they are taking turns blowing each other out. You really see it take effect where there is a significant time change. BC will make the East Division teams play late as Hell, and they will just hit a wall at halftime. Then the East Division teams will make BC play a noon game (EST) and the Lions will get crushed.
Different league, country and rules so not the greatest comparison, but it's the only FB league I know of where you can frequently see the effects of HFA in back to back weeks. And it is a Helluva lot more than 3-6 points.
-
You're going to hate me for this, but I think who you play the week before matters here.
medina, all of what you've said makes sense, and it's likely true.
One thing I've never come across are any interviews with OC or DC asking if their play calls are ever influenced by home vs road. I doubt it, but it may be the case.
-
Extreme weather can enter the fray, too.
The 2014 game vs. Ohio State featured a high of 21 degrees and sideways snow for the duration of the game. It was clear that the Bucks didn't want to be out there and were playing at about 3/4 speed the whole time. I also distinctly remember the number opening at ~17 and not budging despite all the red flags going up. Also, Dilly Bar Dan executed a brilliant piece of psychological warfare.
https://www.sbnation.com/college-football/2014/11/18/7241877/minnesota-ice-cream-sideline-dilly-bar-dan-lehman (https://www.sbnation.com/college-football/2014/11/18/7241877/minnesota-ice-cream-sideline-dilly-bar-dan-lehman)
-
some teams also just flat out play poorly on the road, thus increasing their road opponents HFA
-
Things like weather and who you just played should normalize out of any analysis. It really should be about HFA and nothing else, normalized to take out "Akron".
I wish the Bobs was still around.
I was pondering this thinking about the OSU-UM game, which some refer to as "a game" or something like that. If we presume the teams are close to equal in talent this year, how much advantage should UM be given for the location?
-
I wish the Bobs was still around.
Amen!
-
You're going to hate me for this, but I think who you play the week before matters here.
Eh, we don't hate you for that, we have plenty of other reasons, LoL. Kidding.
I think that who you play the week before definitely matters and that is somewhat related to HFA but I think it is a separate thing.
In the B1G I think you can usually assume that a game against Wisconsin is going to be a physically difficult game. Even when Ohio State has been much better than the Badgers and beaten them soundly, it has still been a physically punishing game so I think you could probably give each UW opponent something like a -3 the following week.
-
As an avid CFL watcher, they do a lot of H-A series in back to back weeks, and seems like it is generally about a 30 point swing. As such I find that HFA is rather underrated.
In CFB we are only able to compare from one season to the next, where they are completely different teams. In the NFL they play the divisional opponents H-A, but rarely if ever in back to back weeks. Teams ebb and flow throughout the year, so it throws off the data a little.
But in the CFL, it is the same two teams playing a game against each other in back to back weeks, and you hardly ever see one team sweep the other. It is almost always a split with the home teams winning, and often times they are taking turns blowing each other out. You really see it take effect where there is a significant time change. BC will make the East Division teams play late as Hell, and they will just hit a wall at halftime. Then the East Division teams will make BC play a noon game (EST) and the Lions will get crushed.
Different league, country and rules so not the greatest comparison, but it's the only FB league I know of where you can frequently see the effects of HFA in back to back weeks. And it is a Helluva lot more than 3-6 points.
The first such back to back H-A series just wrapped up between Hamilton and Montreal.
First Hamilton won their home game 41-10.
One week later Montreal won their home game 36-29.
So a split, and a 38 point swing in point differential, with the only discernable difference being the venue. Not even a time zone change.
-
My own opinion is that it's worth less than "we" think. We see OSU playing at AA and the natural thought is "That could be tough.".
I think it's a factor and varies (Penn State at night) but isn't as large as "we/I" tend to think. Any analysis should discard home games against pastries.
https://www.saturdaydownsouth.com/sec-football/home-field-advantage-sec-college-football/ (https://www.saturdaydownsouth.com/sec-football/home-field-advantage-sec-college-football/)
“(Home-field) is typically worth three points but it can be more for certain venues where a team has been particularly tough at home for a stretch of several years,” Edwards said. “Also, if a team travels across country, especially on a short week (BYU at UCF last week), that can result in adding another point (or more) to the home-field advantage. Weather can be another variable … like a team from up North playing at Miami at noon eastern in September or early-mid October.”
-
The first such back to back H-A series just wrapped up between Hamilton and Montreal.
First Hamilton won their home game 41-10.
One week later Montreal won their home game 36-29.
So a split, and a 38 point swing in point differential, with the only discernable difference being the venue. Not even a time zone change.
HFA isn't the only thing going on here, IMHO. Other factors include motivation and what I will call "surprise".
Motivation:
No matter what anyone says, I think it is difficult to take game prep as seriously as you should when you are preparing for a team that you already beat by 31 points.
Surprise:
The team that won the first game generally doesn't want to make major changes to their game-plan for the obvious reason that it worked the first time. Conversely, the team that lost the first time obviously knows that their game-plan from the first game did not get the job done so they HAVE to try something different. That puts them in a bit of a "nothing to lose" situation where they might as well try something radical.
My least favorite example of this is the 1975 Buckeyes. This was Woody's last great team. Woody's last three teams (1976-1978) each had at least three non-wins. The 1975 team was the fourth in a row to do no worse than 9-2 (9-2 in 1972, 10-0-1 in 1973, 10-2 in 1974, and 11-1 in 1975).
- Preseason #4: Ohio State finished #4 in the final 1974 poll and started at #4 in the 1975 preseason AP Poll.
- 9/8 Poll #3: The Buckeyes didn't play but #2 Bama lost 20-7 to unranked Mizzou so the Buckeyes moved up to #3.
- 9/15 Poll #3: The Buckeyes shutout #11 Michigan State and stayed #3.
- 9/22 Poll #2: The Buckeyes beat #7 Penn State and moved up to #2 on the strength of back-to-back wins over top-11 opponents.
- 9/29 Poll #2: The Buckeyes beat UNC and stayed at #2 (but they were getting VERY close to #1 OU).
- 10/6 Poll #1: The Buckeyes blew out #13 UCLA on the road and moved up to #1.
- 10/13 Poll #1:
- 10/20 Poll #1:
- 10/27 Poll #1:
- 11/3 Poll #1:
- 11/10 Poll #1: Previous #2 OU lost to an unranked Kansas team which had the effect of consolidating Ohio State's hold on #1. In the 11/10 Poll the Buckeyes had 49 of the 58 first place votes.
- 11/17 Poll #1:
- 11/24 Poll #1: A few things happened here: First, Ohio State went to Ann Arbor and beat #4 Michigan 21-14. Second, #2 Nebraska lost to #7 OU. At this point Ohio State had 56 of 59 first place votes and a 150 point lead over #2 aTm. As it turned out, aTm was a bit of a paper tiger. After beating Texas on 11/28 they got blown out by Arkansas (12/6 in Little Rock) and USC (12/22 in the Liberty Bowl).
- 12/1 Poll #1: This was the penultimate poll of 1975 and the Buckeyes were firmly in control with 50 of 58 first place votes and a substantial lead over #2 aTm which was about to lose two straight anyway.
In the final pre-bowl poll the Buckeyes were firmly in command. Then the Rose Bowl happened.
Remember that #13 UCLA team that Ohio State blew out to move up to #1?
The Buckeyes lost to (then) #11 UCLA in a Rose Bowl rematch to finish #4 behind #1 Oklahoma (lost 23-3 to a Kansas team that finished 7-5), #2 ASU (played a WAC schedule), and #3 Bama (ran the table after the aforementioned 20-7 loss to a Mizzou team that finished 6-5).
Plausible contenders in the bowls:
- #1 Ohio State lost 23-10 to #11 UCLA in the Rose Bowl
- #2 aTm lost 20-0 to unranked USC in the Liberty Bowl and also lost 31-6 to #18 Arkansas
- #3 Oklahoma beat #5 Michigan 14-6 in the Orange Bowl
- #4 Bama beat #8 Penn State 13-6 in the Sugar Bowl
- #5 Michigan lost 14-6 to #3 Oklahoma in the Orange Bowl
- #6 Nebraska lost 17-14 to #7 ASU in the Fiesta Bowl
- #7 ASU beat #6 Nebraska 17-14 in the Fiesta Bowl
The Buckeyes didn't even get any first place votes in the final poll despite the fact that their 13 point loss to final #5 UCLA was a heck of a lot better than #1 Oklahoma's 20 point loss to unranked Kansas or #3 Bama's 13 point loss to unranked Mizzou. Oh well.
-
Good analysis, Medina.
I remember that I was rooting for Michigan to beat Ohio State, as I thought that there was more likelihood of that happening than the Buckeyes losing in the Rose Bowl, and OU certainly wasn't going to catch them if they remained unbeaten.
The Buckeyes were unlucky in that their one loss was their last game. Oklahoma's loss to Kansas was a bigger loss to a worse team, but it came in early November.
All other things being equal, Ohio State probably would have been national champs had they lost to UCLA early in the season and then won the Rose Bowl rematch.
OTOH, everything broke right for OU there at the end.
-
You're going to hate me for this, but I think who you play the week before matters here.
medina, all of what you've said makes sense, and it's likely true.
One thing I've never come across are any interviews with OC or DC asking if their play calls are ever influenced by home vs road. I doubt it, but it may be the case.
Who you play the week before matters a helluva lot. I will never forget the perfect example for me. Drew will object because it concerns ut, but perhaps the others won't because it also involves PSU.
When Coach Bryant and Paterno scheduled the 10 year ('81-'90) series, Coach Bryant was compelled to use FOUR dates ('82-'85) he had set aside in order to take a bye before the ut game. From 1971-1994 Bama lost only FOUR games to ut. The FOUR losses were consecutive. Guess which FOUR years in that glorious (for Bama) 24 year period that Bama lost to ut.
http://football.stassen.com/cgi-bin/records/opp-opp.pl?start=1971&end=1994&team1=Alabama&team2=Tennessee (http://football.stassen.com/cgi-bin/records/opp-opp.pl?start=1971&end=1994&team1=Alabama&team2=Tennessee)
-
I'm going to start calling medina "MiniBobs".
-
I don't remember Bobs being around for all that long quite frankly
-
Bobs was a long timer at other previous sites
-
I'm talking from '03/'04 when I joined CFN until moving here.Most of the guys posting here and some who haven't made were more prevelant - IMO
-
The Bobs was around long enough to get a nickname. I recall having some long discussions with him where he used his data base.
-
Bobs wasn't the most chatty guy, small doses but with substance
-
Bobs wasn't the most chatty guy, small doses but with substance
He could be surly at times, little patience for idiots like me. I had to adjust and realize he really KNEW his stuff and when he spoke, it was real.
He hung me out to dry a few times.
Well, so has nearly everyone. I vaguely recall being new here and giving Badger some guff.