The D3 Win Matrix

What you see below is a win matrix of games played through 2/15/24. The teams were placed on the margins in the T100+ order and represent all games played between D3 opponents.  All wins are indicated by black rectangles. Those above the diagonal were won by teams presently ranked higher while those indicated below would be upsets moving forward. At the time most were upsets, but if the same two teams played each other again, the one that lost the first time would be forecasted to avenge that loss.  The reason some win probabilities are now higher relates to subsequent matches played by both teams.  If there hadn’t been any since their match, then the probability for a favorite to win certainly would decrease because of the initial defeat.  The distance below the diagonal indicates how unlikely a repeat dog win would be to happen.  For example another Baruch win over NYU at a neutral court would be very unlikely, given the distance that win is below the diagonal.  Roughly 90% of all wins are above the diagonal, but that doesn’t mean 90% were forecasted properly – 84% actually were. It just means with the benefit of hindsight to have put the teams in this order earlier, it would have produced a 90% forecast rate.

A matrix like this gives not only a visual of who’s defeating whom, but also who is playing whom. Notice the scarcity in the top right? Thank goodness for that. Nobody wants to watch the best 20 teams in the landscape playing the weakest 20, especially themselves! A strong density of black rectangles close to the diagonal indicate similar quality teams scheduling matches with each other, in particular the higher density in the upper left showing really good teams meeting head-to-head. I suspect the density along the diagonal will intensify as conference matches begin to make their way on to the schedule. This is because there is the propensity for conferences to be made up of programs competitive with each other, and they all play no less than a single “round robin” of matches to arrive at their final standings, which doubles as seeding for conference tournaments to follow.

I circled some of the matches already played that are farthest from the diagonal. All but one were 3-2 “upsets,” or at least they would be upsets if they were to go down that way on a neutral court in the future – Based on present ratings, anyway. I am scratching my head a little bit at the Tuesday match when Hiram defeated Mount Union 3-0. Congrats to them for defeating a team that was 37 ranked spots in front of them at the week’s beginning on the T100. They didn’t just pull a rabbit out of a hat with a 3-2 squeaky margin and a negative point differential. No sir! They went 3-0, winning sets by 6, 3, and then 9 points for an 18 point differential sweep. Even IH and Massey had them 33 & 40 positions apart in similar fashion as the T100 did to start the week. The Terriers lifted their hind leg and showed us all what they thought of this consensus opinion as it pertains to them. With Wittenberg, Wabash, and Baldwin Wallace coming up, they have an opportunity to make a serious move well into the top half of all D3 teams in the landscape, and maybe set the stage to separate themselves from what was shaping up as a nip & tuck AMCC race between no less than 5 candidates, one to earn their way to a play date later in April. Often, matches like this are rich with a backstory to help explain such an event. If there is one out there to be heard please do tell! In the meantime the Spotlight match of the week between Naz and North Central is getting ready to start in just 15 minutes as I finish this last thought. For those tuning in, enjoy what hopefully will be a great ride!