Category Archives: ABOVE

2017 WCHA Broadmoor Trophy Projections

Standard ABOVE

  • Michigan Tech: 62% of the time
  • Bowling Green: 38% of the time

Best-Worst Randomization

  • Michigan Tech: 65% of the time
  • Bowling Green: 35% of the time

I’ve been doing ABOVE projections with a randomizing element since late January.  The reason for this is pretty basic: if you want to test that an estimate is right, you can use a group of estimates clustered around a single estimate and see if any of them are right or wrong.  I’ve been using a randomization band of +/- 40, so in a case where I ran 81,000 cases (I run 100,000 to 1,000,000 based on how many games I’m estimating), I’d expect to see around 1,000 uses of each BELOW input.

That first estimate is the randomized BELOW estimate churned through the ABOVE calculation 1,000,000 times.  Coming into the final, Michigan Tech is at BELOW 1701 and Bowling Green is at 1617, its highest mark in 2016-17.  As a result, we could have a matchup of MTU 1741, BGSU 1577; in that matchup, Michigan Tech would be picked to win 72% of the time.  We could also see MTU 1661, BGSU 1657, and that’s a dead heat.

When I noticed that BGSU was at its best mark of the year, I decided to additionally randomize between the best and worst marks for each team.  MTU has been as high as 1739 and as low as 1586; BGSU is at 1617 now and was as low as 1479.  This gives us a band of 54% BGSU favored to 82% MTU.  Throw that in the wash and you get that second pair of numbers.

Which is better?  It’s a single game.  BG fans will probably rib me if their team wins on Saturday.  MTU fans will probably accuse me of jinxing them.  I’ll probably be frustrated either way, but it’s a certainty that I’ll be back next year.

[Go Tech]

Ed.: Post updated to fix the date of the final game.

Semifinal and Final Projections, 2017-03-12

Congratulations to Bowling Green for downing top seed Bemidji State, a team that hadn’t lost two league games in a row all season and lost both games in a series just twice (to #1 North Dakota and a streak-starting Princeton team).

Minnesota State pulled off a 1-0 win over Michigan Tech to push the 2-3 semifinal to a third game.  Here are your projections:

  • Michigan Tech is expected to win on Sunday 58.131% of the time. If they win, they are favored to beat Bowling Green in Houghton next weekend 67.802% of the time.
  • In the 41.869% of trials where Minnesota State wins on Sunday, they beat BG in a game they host 60.981% of the time.
  • Despite the fact that they are a strong underdog in both finals matchups, Bowling Green is projected to be the Broadmoor Trophy winner 35.054% of the time, second to Michigan Tech at 39.414% and ahead of Minnesota State at 25.532%.

If Michigan Tech wins on Sunday and Bowling Green wins next week, we will have the same combination of playoff seed wins as last year: 1-2-3-4 in the quarters, then 2-4 in the semis with a 4 in the finals.  It would be truly interesting for the Falcons to win given their start.

Enjoy the games!  We have just two left, and that is sad.

Updated Projections, 2017-03-11

After a 5-1 Michigan Tech and a 4-3 overtime Bowling Green win, it’s time to update Friday’s projections.

Overall Broadmoor Trophy Projection:

  • Bemidji State: 16.278%
  • Michigan Tech: 48.734%
  • Minnesota State: 11.848%
  • Bowling Green: 23.140%

Bowling Green (1587) at Bemidji State (1640)

  • Bowling Green sweep: 21.618%
  • Bowling Green in 3: 42.469%
  • Bemidji State in 3: 35.91%

Minnesota State (1642) at Michigan Tech (1700)

  • Michigan Tech sweep: 58.282%
  • Michigan Tech in 3: 21.428%
  • Minnesota State in 3: 20.290%

Michigan Tech at Bemidji State final:

  • Occurs 28.525% of the time
  • Bemidji State wins 44.673% of those matchups.
  • Michigan Tech wins 55.327% of those matchups.

Minnesota State at Bemidji State final:

  • Occurs 7.332% of the time
  • Bemidji State wins 48.213% of those matchups.
  • Minnesota State wins 51.787% of those matchups.

Bowling Green at Michigan Tech final:

  • Occurs 51.053% of the time
  • Michigan Tech wins 64.545% of those matchups.
  • Bowling Green wins 35.455% of those matchups.

Bowling Green at Minnesota State final:

  • Occurs 13.091% of the time
  • Minnesota State wins 61.508% of those matchups.
  • Bowling Green wins 38.492% of those matchups.

 

2017 WCHA Semifinal and Final Predictions

First semifinal: Bowling Green (BELOW 1568) at Bemidji State (BELOW 1659)

  • Bemidji wins 66.850% of the time: in 2: 41.926%; in 3: 24.924%
  • BG wins 33.150% of the time: in 2: 16.591%; in 3: 16.559%

Second semifinal: Minnesota State (BELOW 1672) at Michigan Tech 1670

  • Tech wins 49.614% of the time: in 2: 27.742%; in 3: 21.871%
  • Mankato wins 50.386% of the time: in 2: 22.109%; in 3: 28.277%

Finals Matchups:

  • Bemidji-Tech: 33.145%.
  • Bemidji-Mankato: 33.705%.
  • Tech-BG Final: 16.460%.
  • Mankato-BG Final: 16.681%.

Broadmoor Champions

  • Bemidji State: 33.350%
  • Michigan Tech: 24.850%
  • Minnesota State: 25.242%
  • Bowling Green: 16.558%

Once you get to the finals, all four are 50-50 matchups.  Here’s why:

  1. Because Michigan Tech and Minnesota State are in a dead heat with Bemidji State close behind, the Tech-Mankato winner is going to be just a little bit better because that season is generally going to be close.
  2. If Bowling Green gets the upset, they’ll be pretty close to their finals opponent.

This is a lot like last year, where the top three teams were pretty clustered in terms of an estimate and Ferris broke through and were the better team that weekend.  ABOVE, by randomizing BELOW estimates within a range, accounts for teams getting hot and going cold and, over time, averages that out.  Now it makes more sense for a regular season projection, but it works here in the finals.

So what you see above is the projection of Bemidji State likely winning 2/3 of the time and the Tech-Mankato semifinal being a toss-up.  I could run that 1,000,000,000 times and probably get something close to the same result as I did with 1,000,000.

Enjoy the games.  The math means little when the puck drops.

2017 WCHA Playoffs Quarterfinal Projections

I haven’t re-written all of the code (gotta love being on-call), but here are projections for the four quarterfinals and the combinations that result:

Quarterfinal #1: Northern Michigan at Bemidji State

  • Bemidji sweeps 68.587% of the time and wins in three 20.975% of the time.
  • Northern sweeps 4.335% of the time and wins in three 6.103% of the time.

Quarterfinal #2: Lake Superior at Michigan Tech

  • Tech sweeps 65.791% of the time and wins in three 21.993% of the time.
  • Lake State sweeps 5.191% of the time and wins in three 7.025% of the time.

Quarterfinal #3: Alaska at Minnesota State

  • Mankato sweeps 51.005% of the time and wins in three 24.905% of the time.
  • Fairbanks sweeps 11.244% of the time and wins in three 12.846% of the time.

Quarterfinal #4: Ferris State at Bowling Green

  • BG sweeps 33.769% of the time and wins in three 24.982% of the time.
  • Ferris sweeps 20.526% of the time and wins in three 20.722% of the time.

Semifinal matchups

  • Bemidji – Tech – Mankato – BG : 35.047%
  • Bemidji – Tech – Mankato – Ferris: 24.603%
  • Bemidji – Tech – BG – Fairbanks: 11.138%
  • Tech – Mankato – BG – Northern:  9.278%
  • Bemidji – Tech – Ferris – Fairbanks:  7.818%
  • Bemidji – Mankato – BG – :  4.862%
  • Bemidji – Mankato – Ferris – Lake State:  3.453%
  • Tech – Mankato – Ferris – Northern:  2.864%
  • Bemidji – BG – Fairbanks – Lake State:  1.557%
  • Tech – BG – Fairbanks – Northern:  1.288%
  • Bemidji – Ferris – Fairbanks – Lake State :  1.083%
  • Tech – Ferris – Fairbanks – Northern:  0.908%
  • Mankato – BG – Lake State – Northern:  0.566%
  • Mankato – BG – Lake State – Northern:  0.398%
  • BG – Fairbanks – Lake State – Northern:  0.175%
  • Ferris – Fairbanks – Lake State – Northern:  0.121%

Methodology

This is pretty standard ABOVE stuff:

  1. Take the final 2016-17 BELOWs and randomize them +/- 40 points to enhance their quality as estimates.
  2. Calculate expected values using the Elo-style formulas.
  3. Have a random dice roll for an outcome of the game.
  4. Determine a winner of the game based on the expected values.
  5. Re-calculate BELOW (based on the randomized estimates).
  6. Repeat steps 2-4.
  7. If one team has one two games, they’ve swept and are declared the winner.
  8. If the teams split the first two games, repeat steps 2-5 to get the third game.
  9. Perform steps 7 and 8 for each of the four quarterfinals and account for one of the sixteen possible semifinal matchups.

I just didn’t get to finishing the semifinal and final codes for this.  I’ll look to do that tomorrow when perhaps I’ve slept.

2016-17 WCHA Final Playoff Projections

TeamSeed  
Bemidji State1 - 100%
Michigan Tech2 - 100 %
Minnesota State3 - 100 %
Bowling Green4 - 65.91%5 - 34.09%
Ferris State4 - 34.09%5 - 65.91%
Alaska6 - 100%
Lake Superior7 - 84.25%8 - 14.60%9 - 1.16%
Northern Michigan7 - 14.60%8 - 53.85%9 - 31.56%
Alabama-Huntsville7 - 1.16%8 - 31.56%9 - 67.29%
Alaska-Anchorage10 - 100%

I’d like to thank the league for putting out a press release so I didn’t have to work on this at work, and I’d like to thank the teams for making it easy on me — we know half of the results already.  Here’s what’s left:

  1. Bowling Green and Ferris State are tied at 41 points, and the Falcons own the tiebreaker.
  2. Lake Superior is at 8-3-3-13/33, Northern Michigan is at 9-1-2-15/31, and Alabama-Huntsville is at 9-0-3-15/30.
  3. Lake Superior makes the playoffs as the 7th seed if they get a win tonight in regulation or overtime.  They also get 7th as long as they have more points than Northern Michigan, who wins at B.
  4. Northern Michigan makes the playoffs at the 7th seed if they pass or equal Lake Superior in the standings unless Alabama-Huntsville  ties the two at 33, which takes the Lakers losing and Northern winning in 3×3 or the shootout and the Chargers winning over Bowling Green.
  5. Lake Superior finishes in 8th if Northern passes or ties them without Huntsville doing the same.  UAH winning the three-way tiebreaker is the only way that LSSU misses the playoffs.
  6. Northern Michigan finishes in 8th anytime they stay ahead of Huntsville but don’t pass Lake State.
  7. The NMU-UAH tiebreaker is fun: at 33, UAH wins at B; at 32, at D; at 31, at C.  That’s just wacky.

 

2016-17 WCHA Week 21, Series 70 Projection: Northern Michigan (1528) @/v Michigan Tech (1666)

2016-17 Week 21, Series 70: Michigan Tech (1666) at/home Northern Michigan (1488)

ResultProbability
MTU Sweeps40.64%
Splits27.17%
NMU sweeps8.20%
MTU 5 points (W, OTW)7.71%
MTU 4 points (W, OTL or OTW, OTW)5.05%
NMU 5 points (W, OTW)3.39%
NMU 4 points (W, OTL or OTW, OTW)2.52%

This is our lop-sided matchup of the week, and it’s a shame, really.  The Wildcats’ resurgence has been remarkable, but facing the league’s best team, even on rivalry weekend, is a tough task.  They have to like their chances of keeping ahead of the pack with a split or better — about 41% — but the long losing streak they had has made the playoffs a dicey proposition.

2016-17 WCHA Week 21, Series 68 Projection: Alabama-Huntsville (1413) at Bowling Green (1479)

2016-17 Week 21, Series 68: Alabama-Huntsville (1413) at Bowling Green (1479)

ResultProbability
Splits30.99%
BGSU Sweeps30.47%
UAH sweeps14.54%
BGSU 5 points (W, OTW)6.57%
UAH 5 points (W, OTW)4.64%
BGSU 4 points (W, OTL or OTW, OTW)4.13%
UAH 4 points (W, OTL or OTW, OTW)3.39%

This series projection is probably a lot closer than people would expect, but BELOW likes UAH’s win over BSU a lot and brings them in, and it looks at the Falcons going 6-0-0-10 in their last 16 WCHA games and gives them a big ol’ ¯(°_o)/¯.

2016-17 WCHA Week 21, Series 67 Projection: Minnesota State (1644) at Bemidji State (1662)

2016-17 Week 21, Series 67: Minnesota State (1644) at Bemidji State (1662)

ResultProbability
Splits32.13%
BSU Sweeps23.99%
MSU sweeps23.99%
BSU 5 points (W, OTW)5.69%
MSU 5 points (W, OTW)5.46%
MSU 4 points (W, OTL or OTW, OTW)4.02%
BSU 4 points (W, OTL or OTW, OTW)3.53%

That’s as even as even gets, and series like these are why I put the randomization into the model — BELOW is an estimate, and if I can randomize that estimate in a range, I stand a better shot at getting a more realistic projection given lots of variability.  ABOVE can shift the model up to 80 points in each direction, which would put the Beavers up near their heights while the Mavericks would be down near their season average.

2016-17 WCHA Week 18, Series 58 Prediction: Northern Michigan (1426) at Bowling Green (1533)

2016-17 Week 18, Series 58: Northern Michigan (1426) at Bowling Green (1533)

ResultProbability
BGSU sweeps50.59%
Splits27.98%
BGSU 5 points (W, OTW)8.99%
NMU Sweeps8.91%
BGSU 4 points (W, OTL or OTW, OTW)6.66%
NMU 5 points (W, OTW)3.82%
NMU 4 points (W, OTL or OTW, OTW)2.05%

Before NMU smoked UAA — a team that’s no great shakes this year, but a team that BELOW considered to be on par with the Wildcats — it looked like Northern Michigan might be headed for the scrap heap.  Instead, its strong showing against Alaska-Anchorage and the struggles of Alabama-Huntsville could take Walt Kyle off of the hot seat in Marquette.

Chris Bergeron’s team has to find an identity.  Minnesota State has a date against cellar-dweller UAA (who admittedly are a tough out, especially on Friday nights at home), and while the two teams immediately behind them are idle this weekend, the fact of the matter is that BG plays just one series after this weekend — home against UAH to close the season.  It’s put-up or shut-up time at the Mad House on Mercer for a team with just five games to play this month.