×
Rock Chalk Talk: Basketball
Anything pertaining to basketball: college, pro, HS, recruiting, TV coverage
Anything pertaining to basketball: college, pro, HS, recruiting, TV coverage
predictions for Duke game
- asteroid
- Topic Author
- Offline
- Platinum Member
Less
More
- Posts: 600
- Thank you received: 3139
2 years 1 week ago #29770
by asteroid
Predicting today's game is difficult, because there is not much to go on. Both
teams have won two home games against cupcakes. Both teams will be without their
Hall of Fame coach, though for different reasons.
The more reliable of the prognosticators have it as a close game, a one-possession
affair, meaning it would be easy to go either way.
Dolphin is now on board for this season, as is Colley, though it appears that
Colley doesn't use any bias from last season, as his top teams are San Diego
State, Mississippi State, and Nevada. Duke is way down at #45, and Kansas is
even lower at #71, according to Colley.
Real Time is on board for this season, but Real Time still isn't ready for Prime
Time. Once again, the "Scouting" prediction of 73 to 84 is inconsistent with the
Gamer prediction of 71 to 84, where one can see that each team has a record of 1-0,
while at the same time having a home record of 2-0! Also, Real Time has the game
at Duke, not at a neutral site. No wonder Real Time has the most pessimistic
prediction, with Kansas losing by 13 points, especially considering his past
practice of using a large home court advantage.
Dunkel isn't quite as bizarre, but when you have a game between a #2 ranked
Team A and a #6 ranked Team B on a neutral court, you would expect Team A to
be favored. Nope. Dunkel is picking Team B to win by 2 points. Go figure.
Oh, and Team B in this case is Duke.
Seven Overtimes has rankings for this season, but all 51 games today have
predicted scores of 60-60, obviously a placeholder, not a real prediction.
Aside from Real Time, the most optimistic and the most pessimistic predictions
both come from Sagarin. His Recent Games method favors Duke by 5, whereas his
peculiar eigenvector analysis favors Kansas by 6. The Overall, Predictor, and
Golden Mean methods all have margins of less than 2 points, with Overall and
Predictor favoring Duke, and Golden Mean favoring Kansas.
Kansas has the #300 strength of schedule, while Duke is #329. I don't like
seeing numbers that low. Starting the season with the Champions Classic
tended to keep those numbers lofty, even after some cupcakes followed. At
least we'll see a boost in those numbers in tomorrow's updated ratings.
Duke has allowed at most 44 points against its two opponents, averaging 41
points, while the Kansas defense has given up 20 points more than that on
average. Then again, the Jayhawks really lifted the foot off the accelerator
in the second half against North Dakota State. Still, Pomeroy has Kansas with
the #10 adjusted defense, while Duke has the #33 adjusted defense. On the
flip side, Pomeroy has Duke with the #4 adjusted offense, while Kansas is only
#14. That suffocating Blue Devil defense has kept their games to 118.5 total
points on average, and that is keeping the total points prediction for today's
game on the low side. Some of the prognosticators don't think either team
will score 70 points or more, though Massey and Vegas have the two teams
scoring in the low to mid 70s.
Rock Chalk!
Greenfield Stats Comparison
===========================
Offensive Stats Duke KU Defensive Stats Duke KU
Points/Game 77.5 85.5 Opp Points/Game 41.0 61.5
Avg Score Margin +36.5 +24.0 Opp Effective FG % 35.5 43.5
Assists/Game 17.5 23.0 Off Rebounds/Gm 16.0 14.0
Total Rebounds/Gm 48.0 44.5 Def Rebounds/Gm 29.0 29.0
Effective FG % 51.6 56.0 Blocks/Game 5.0 5.5
Off Rebound % 47.1 39.4 Steals/Game 10.5 8.0
FTA/FGA 0.291 0.177 Personal Fouls/Gm 12.0 14.5
Turnover % 15.0 10.5
My Stats Comparison KU Duke
=================== ============= ============
performance -0.98 +9.95
inconsistency 3.34 8.55
trend +4.73 ± . +12.09 ± .
mental toughness +0.70 ± . -1.75 ± .
average total pts 147.00 118.50
Common Opponents
================
There are no common opponents.
Players to Watch
================
most minutes Jeremy Roach (guard)
most points Mark Mitchell (forward)
most rebounds Kyle Fillpowski (center)
most assists Jeremy Roach (guard)
most steals three-way tie
most blocks Dereck Lively II (center)
most turnovers Jeremy Roach (guard)
most fouls two-way tie
Forward Dariq Whitehead is out with a right foot fracture.
2-0 2-0
Margin Predicted Win Prb Kansas Duke
Predictor points Score percent Rating SOS Rating SOS
==================== ====== ========= ======= ============ ============
Sagarin Overall -1.80 63 65 44 # 9 #300 # 4 #329
Sagarin Predictor -0.22 64 64 49.2 # 5 #300 # 4 #329
Sagarin Golden Mean +0.92 65 64 # 3 #300 # 5 #329
Sagarin Recent Games -4.99 62 67 # 15 #300 # 3 #329
Sagarin Eigenvector +5.91 67 71 69
Massey +2.00 75 73 56 # 2 #193 # 6 #246
Pomeroy +0.22 72 72 # 10 #319 # 11 #306
Greenfield +1.00 72 71 # 15 #255 # 6 #172
Dunkel -2.00 68 70 # 2 # 6
Vegas (via Dunkel) -1.50 72 74
Dolphin Predictive -2.49 65 67 41.5 # 4 #233 # 1 #257
Real Time -13.00 71 84 21.9 # 5 #206 # 3 #167
Seven Overtimes # 29 #269 # 71 #339
DPPI # # # #
ESPN BPI +1.60 56.1 # 13 #339 # 23 #267
Whitlock *** +2.67 # # # #
Colley Matrix -1.21 # 71 #326 # 45 #283
NCAA NET # #
LRMC # # # #
common opponents
Massey composite # 5 # 6
Pomeroy offense # 14 # 4
Pomeroy defense # 10 # 22
Pomeroy tempo #117 #239
==================== ====== ========= ======= ============ ============
average -0.86 68.0 70.2 48.2
scatter 4.23 4.3 5.6 14.8
Here is Kansas' season, for which the current season projection is a 22-9
record, assuming the higher rated teams advance in the Battle 4 Atlantis:
SITE OPPONENT SCORE PREDIC ERROR PROB.
---- ----------------------------- ------- ------ ----- -----
HOME #311 Omaha(Neb.-Omaha) 89 64 +28.34 -3.34
HOME #195 North Dakota State 82 59 +21.61 +1.39
NEUT # 4 Duke -0.22 0.492
HOME #206 Southern Utah +22.26 0.978
NEUT # 48 NC State +6.91 0.735
----- one of the following two games -----
NEUT # 25 Wisconsin +3.85 0.637
NEUT # 44 Dayton +6.68 0.728
----- one of the following four games -----
NEUT # 29 Tennessee +4.45 0.657
NEUT # 51 Butler +7.32 0.747
NEUT # 55 Southern California +7.94 0.765
NEUT # 56 BYU +8.07 0.768
-----
HOME #214 Texas Southern +22.72 0.981
HOME # 32 Seton Hall +7.95 0.765
AWAY # 89 Missouri +8.67 0.785
HOME # 6 Indiana +3.29 0.618
HOME #154 Harvard +19.27 0.960
HOME # 39 Oklahoma State +8.92 0.791
AWAY # 22 Texas Tech +0.38 0.514
AWAY # 33 West Virginia +2.10 0.576
HOME # 69 Oklahoma +12.48 0.872
HOME # 37 Iowa State +8.81 0.788
AWAY # 61 Kansas State +5.61 0.695
HOME # 95 TCU +15.07 0.915
AWAY # 2 Baylor -5.19 0.319
AWAY # 8 Kentucky -2.67 0.404
HOME # 61 Kansas State +11.77 0.858
AWAY # 37 Iowa State +2.65 0.595
HOME # 10 Texas +4.46 0.657
AWAY # 69 Oklahoma +6.32 0.717
AWAY # 39 Oklahoma State +2.76 0.599
HOME # 2 Baylor +0.97 0.535
AWAY # 95 TCU +8.91 0.791
HOME # 33 West Virginia +8.26 0.774
HOME # 22 Texas Tech +6.54 0.724
AWAY # 10 Texas -1.70 0.439
Here is Duke's season-to-date:
SITE OPPONENT SCORE PREDIC ERROR PROB.
---- ----------------------------- ------- ------ ----- -----
HOME #216 Jacksonville 71 44 +23.09 +3.91
HOME #325 USC Upstate 84 38 +30.00 +16.00
NEUT # 5 Kansas +0.22 0.508
The following user(s) said Thank You: sasnak, hairyhawk, Bayhawk, boulderhawk, Socalhawk, porthawk, jaythawk1
Please Log in or Create an account to join the conversation.
Share this page:
- USAF Jayhawk
- Offline
- Elite Member
Less
More
- Posts: 291
- Thank you received: 314
2 years 1 week ago #29779
by USAF Jayhawk
Not that it matters now, but where is the actual prediction?
Please Log in or Create an account to join the conversation.
- asteroid
- Topic Author
- Offline
- Platinum Member
Less
More
- Posts: 600
- Thank you received: 3139
2 years 1 week ago #29780
by asteroid
There were a bunch of predictions, the average of which was Kansas 68, Duke 70.
But now I see a typo in the table. Sagarin's eigenvector analysis, which came the closest to the actual margin, should have been Kansas 67, Duke 61, not Duke 71. The Sagarin totals should have all been 128 points, and the eigenvector analysis was for a win, but the scores shown didn't reflect the win. I'll chalk that error up to the fact that I was short on time, needing to get to the airport for a flight. Anyway, if you take those 10 points away, the average for Duke would have decreased only by a little over a point, so the average of the predictions would have been more like Kansas 68, Duke 69, still a loss. But drop RealTime from the analysis, and we're looking at a toss up.
My own personal prediction was Kansas 67, Duke 66, though not posted here. I got the point total bang on. Just a bucket off in each direction.
But now I see a typo in the table. Sagarin's eigenvector analysis, which came the closest to the actual margin, should have been Kansas 67, Duke 61, not Duke 71. The Sagarin totals should have all been 128 points, and the eigenvector analysis was for a win, but the scores shown didn't reflect the win. I'll chalk that error up to the fact that I was short on time, needing to get to the airport for a flight. Anyway, if you take those 10 points away, the average for Duke would have decreased only by a little over a point, so the average of the predictions would have been more like Kansas 68, Duke 69, still a loss. But drop RealTime from the analysis, and we're looking at a toss up.
My own personal prediction was Kansas 67, Duke 66, though not posted here. I got the point total bang on. Just a bucket off in each direction.
Please Log in or Create an account to join the conversation.
- hairyhawk
- Offline
- Platinum Member
Less
More
- Posts: 1202
- Thank you received: 692
2 years 1 week ago #29783
by hairyhawk
Very nice. Thanks!
Please Log in or Create an account to join the conversation.
- USAF Jayhawk
- Offline
- Elite Member
Less
More
- Posts: 291
- Thank you received: 314
2 years 1 week ago #29786
by USAF Jayhawk
My real point was the two columns that had the various scores were not labeled by team, so I could not tell which score went with KU and which score went with Duke...
Please Log in or Create an account to join the conversation.
- asteroid
- Topic Author
- Offline
- Platinum Member
Less
More
- Posts: 600
- Thank you received: 3139
2 years 1 week ago #29789
by asteroid
Right you are! I've been doing it that way for years, and you're the first to point it out. I've always listed Kansas on the leftmost of those two columns with the opponent rightmost of those two columns. But it can be reverse-engineered by looking at the margin, which has always been given from the Jayhawk perspective. If it's a negative margin, then Kansas will have the smaller of the two scores, assuming no typos on my part.
Please Log in or Create an account to join the conversation.