:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/28695509/157597741.0.jpg)
"Once is happenstance. Twice is coincidence. Three times, it's enemy action." - Auric Goldfinger
This week, your friends at Hustle Belt have pored through 10 years of MAC tournament results to examine the age old question of what happens when teams meet for the third time in the conference tournament. Color commentators like to opine that "it's hard to beat a team three times," but we weren't convinced.
As in last week's analysis, these data cover MAC tournaments from 2004 through 2013. In those 10 tournaments, there were 54 games pitting divisional rivals against each other for the third time in a season. Note that 2014 offers the first chance for non-divisional rivals to play a third time, as could happen for example with Ohio and Toledo (this is called dramatic foreshadowing), because of the change to an 18-game conference schedule.
In those 54 third-time matchups, the favored team won 37 (69 percent) times. But let's look deeper at some other factors that could reasonably affect the outcome.
Does it matter how different the teams are? We looked at the outcomes based on the difference in tournament seeding between the two teams.
Seeding difference of at least 5 spots à favorite won 14 of 21 (67 percent)
Seeding difference of 3 - 4 spots à favorite won 14 of 17 (82 percent)
Seeding difference of 1 - 2 spots à favorite won 9 of 16 (56 percent)
Our hypothesis was that regular season outcomes would best predict the game outcome. And it turns out that this is largely correct, and not consistent with the old commentator aphorism about beating a team thrice.
If the favorite swept the underdog in the regular season à favorite won 21 of 25 (84 percent)
If the underdog swept the favorite in the regular season à favorite won 2 of 3 (67 percent)
If the two teams split the regular season games à favorite won 14 of 26 (54 percent)
What about the total scores posted in the two regular season games? These data turn out to be fairly predictive of outcome as well, which came as a mild surprise.
If the underdog totaled at least 10 more than the favorite à favorite won 2 of 5 (40 percent)
If the underdog totaled 1 - 9 more than the favorite à favorite won 4 of 9 (44 percent)
If the two teams were even à favorite won 1 of 2 (50 percent)
If the favorite totaled 1 - 9 more than the underdog à favorite won 10 of 15 (67 percent)
If the favorite totaled 10 - 19 more than the underdog à favorite won 8 of 10 (80 percent)
If the favorite totaled at least 20 more than the underdog à favorite won 12 of 13 (92 percent)
Simplified, these can be re-summarized as:
Underdog even or better than favorite in total points à favorite won 7 of 16 (44 percent)
Favorite outscores underdog by 1 - 9 points à favorite won 10 of 15 (67 percent)
Favorite outscores underdog by at least 10 points à favorite won 20 of 23 (87 percent)
Finally, a note on the 26 occasions when the third meeting followed a regular season split. It turns out that the team that won the second of the two meetings went on to win 16 (62 percent) of the tournament games, as compared to 10 (38 percent) wins for the team that won the first game.
Now let's get out of the past and steer this ship forward. What the MAC tournament might look like if played today:
Tiebreaker notes:
- Ball State remains seeded ahead of Central Michigan based on their single win being against a better opponent (Buffalo > Kent State).
- Among the quartet of teams at 5-7 in conference, Bowling Green is the last as they've lost to all three others. Northern Illinois split with Kent State and beat Miami, putting them at the top of the group. While Kent State and Miami haven't played yet, the split with NIU gives the Flashes the edge over the ‘Hawks, and a first round home game.
- Ohio's head-to-head win at Buffalo Saturday gives them the tiebreaker, and the all-important #4 seed - and double-bye.
- Finally at the top, Western Michigan has beaten both Akron and Toledo, and is your new leader in the clubhouse. Toledo's win over Akron keeps the Rockets in the #2 seed, and drops Akron to No. 3.
MONDAY
No. 12 Central Michigan @ No. 5 Buffalo
2014 head-to-head: The Bulls won 79-70 in Mount Pleasant on February 8th.
Prediction: Buffalo's home mystique got tarnished a bit this weekend by yet another Ohio second half comeback, but they'll handle the Chippewas in the tournament.
No. 11 Ball State @ No. 6 Eastern Michigan
2014 head-to-head: The Eagles won in Muncie 73-62 last weekend. They meet in Ypsilanti on March 4th.
Prediction: Hard to pick against EMU after watching their demolition of Toledo this weekend, especially given how Da'Shonte Riley should be able to control Majok Majok - and apparently just about anyone else - in the paint.
No. 10 Bowling Green @ No. 7 Northern Illinois
2014 head-to-head: The Huskies went to Bowling Green and won 45-36 on the first weekend of conference play.
Prediction: We know they have the same in-conference record, but Northern Illinois has been the much more complete and consistent team. Huskies move on.
No. 9 Miami @ No. 8 Kent State
2014 head-to-head: These two meet twice in the next three weeks, starting this Wednesday.
Prediction: A tough call given that we worked so hard to crush those stats about third time rematches, and these two give us no data to work with. Kent State does sort of have momentum, and they host the ‘Hawks second, so are more likely to come in with that win. Miami seems to be fading slightly, even though we think they'll actually end up above KSU in the standings. If this game comes to pass, we think Kent State pulls it out.
WEDNESDAY
No. 8 Kent State vs. No. 5 Buffalo
2014 head-to-head: Buffalo won 71-60 at home on January 18th. Rematch at Kent next weekend.
Prediction: Buffalo has to win this game - the much more complete team, even though inconsistency away from home is a concern.
No. 7 Northern Illinois vs. No. 6 Eastern Michigan
2014 head-to-head: None yet, and we're really looking forward to the two games in the next two weeks.
Prediction: Did we mention that the Eagles looked awesome in crushing Toledo? They're primed for the postseason with their upperclassmen and their tough schedule. Honestly we were planning to pick an NIU upset in this round. And we would have over anyone else. Not EMU. Eagles move on.
THURSDAY
Ni. 5 Buffalo vs. No. 4 Ohio
2014 head-to-head: Ohio's second half comeback this weekend gave them a 73-70 win on the road at Buffalo.
Prediction: It's hard to see Buffalo making a run on the road, and obviously Ohio beating them on the road strongly implies the Bobcats' superiority.
No. 6 Eastern Michigan vs. No. 3 Akron
2014 head-to-head: Akron swept the regular season home-and-home with wins on January 22nd and February 5th.
Prediction: We'd love to pick the upset, but look at the previous Q-gazing on coaching success under Keith Dambrot, and then above at the numbers for favored teams that swept in the regular season. Zips move on.
FRIDAY
No. 4 Ohio vs. No. 1 Western Michigan
2014 head-to-head: Western Michigan turned on the offensive gas in the second half and pulled away from Ohio 90-74 in early February. They meet again next weekend.
Prediction: You could correctly say that either of these teams is the best in the MAC - when they play to their potential. Both have been amazingly irregular - even WMU's two wins this week came with fantastical scoring droughts in the first half. We'll take Ohio with the upset here, spurred on by likely a partisan crowd in Cleveland.
No. 3 Akron vs. No. 2 Toledo
2014 head-to-head: Toledo stomped Akron at the JAR 75-61 on January 18th.
Prediction: These have been the two frontrunners through much of conference play, but they both look beatable. Toledo's performance against Akron on the road showed that the Rockets can compete physically inside with the Zips, allowing their explosive guards to create on offense. It's tempting to pick an Akron - Ohio final again, but Toledo advances.
SATURDAY
No. 4 Ohio vs. No. 2 Toledo
2014 head-to-head: Ohio won in overtime at home on February 1st, and then Toledo defended its court with a home win on February 12th.
Prediction: All the data above would suggest that Toledo has a 55 - 65 percent chance of winning this game. But fun fact: the lower seeded team has won half of the last ten MAC championship games. We're taking the upset here - and already looking for extra tickets. Bobcats win in a thriller.
