Week 10 NFL Best Bets now available including the first 4-Star of the season as well as two 3-Stars!
Week 9 and Season Recap
The number of tough losses that my NFL handicapper Mike O’Connor has taken this season is getting ridiculous. Mike had a winning week at 4-3 on his Best Bets, all 2-Star Best Bets, but he had the right side in the 3 losses while the wins were all undoubtedly easy wins.
Let’s begin with the wins. Mike had Minnesota +10 against Dallas and the Cowboys needed a late touchdown just to win the game 27-23 while the Vikings outgained Dallas 393 yards at 5.8 yards per play to 350 yards at 5.6 yppl but were -1 in turnovers. Easy win. No doubt the right side.
The Jets (+6) beat the Saints straight up 26-20 while gaining 340 yards at 6.1 yppl to the Saints 408 yards at 6.3 yppl. New York was +2 in turnovers (both Saints interceptions) but New Orleans is likely to throw more picks when they throw 50 passes while the Jets needed to throw just 20 passes. The Jets were a solid play.
Philadelphia as a dog against the Raiders was obviously a great play, as the Eagles won 49-20. The Raiders got a lot of yards in garbage time to even out the total yards but yards per play is a better indication of a team’s success and Philly averaged 10.1 yppl (after taking out the kneel downs at the end of the game) while the Raiders averaged 6.1 ypp.
Tampa Bay as a huge dog led most of the game, and by as much as 21-0, before losing in overtime and were only outgained by 65 yards. Another great play.
Losses (none of which were bad bets):
The Thursday night loss on Cincinnati -3 was a game where the better team lost 20-22 in overtime as a direct result of turnovers, which are vastly unpredictable. The Bengals were -3 in turnovers as Andy Dalton, who averages less than 1 interception per game in his career, threw 3 picks. The first one set up a Dolphins field goal. The second one came after a long drive got the Bengals to the 10 yard line and that interception was returned 94 yards for a touchdown, a 10 point swing. The Bengals continued to dominate and fought back by scoring on their next two drives to tie the game at 17. On their next drive, in field goal range, Dalton threw his third interception. Cincy’s fumble and Miami’s fumble both led to no scoring but Cincinnati’s 3 interceptions cost them at least 13 points (the 10 point swing on the 94 yard return plus Miami’s FG after the first pick) and the Bengals lost on a safety in overtime despite amassing over 500 yards of total offense. Cincinnati was the right side in that game.
Mike’s second loss was on Buffalo (+3) and the Bills were a very good bet. Buffalo outgained Kansas City 470 yards at 6.1 yppl to 213 yards at 4.2 yppl but were -3 in turnovers. The Bills were so dominant in this game that they could have overcome the -3 in turnovers and covered anyway if not for two of those turnovers being returned for touchdowns. One was with the Bills up 10-3 and on the one yard line when Jeff Tuel threw a pass that was intercepted and returned 100 yards and the other was a fumble return for a touchdown. Buffalo was a great bet.
The third loss was tough to swallow too, as Houston +2 was dominating Indianapolis 21-3 at the half. Houston’s coach Gary Kubiak collapsed heading off the field at the half and was rushed off to the hospital. Normally losing a head coach wouldn’t be that big of a deal, but Kubiak is also the Texans’ play caller and the offense disappeared in the second half without him calling the plays. Houston still outgained the Colts 483 yards at 7.2 yppl to 314 yards at 5.4 yppl but the Texans missed 3 field goals and ultimately lost by 3 points as a 2 point dog. That was another good bet that would win most of the time.
Mike made 7 good bets in week 9 and only went 4-3. This season had been unlucky for Mike heading into this week with numerous close losses and two games he lost that he absolutely should have won (Houston over Seattle and Green Bay over Baltimore) and today added to what has been an incredibly unlucky season. Good bets do lose, as variance with turnovers and special teams is part of football, but that variance is supposed to go both ways and that has not been the case so far this season. I have to go all the way back to week 2 to find a game that Mike won that he really didn’t deserve to win.
I have always kept a ledger to grade my Best Bets honestly and over time my graded percentage is the same as my actual winning percentage because the luck tends to even out over a longer time period. However, there are seasons in which variance plays a role in making a handicapper's record considerably better than it should be or considerably worse. I’m not handicapping the NFL personally this season but I keep a log of Mike O’Connor’s NFL Best Bets and grade those games too. You probably already realize this if you’ve been paying attention, but Mike’s record should be much better than 19-18 given the bad luck he’s had this season, as his handicapping has been just as good as it’s been over his 8 years hitting 58% while his results this year have been heavily influenced by negative variance.
I have 6 categories for grading games. Most games are deserved wins or deserved losses, which are games that turn out the way that they should based on the stats. There are games that are considered toss up wins and toss up losses, which are games that could have gone either way based on the stats or how the game played out. Then there are undeserved wins and undeserved losses, in which the outcome was decided purely by variance. Undeserved wins are games that shouldn't have won but did and undeserved losses are games that should have won and didn't.
So far this season Mike has just 1 win that he didn't deserve and he has 5 clearly undeserved losses, where he had the right side based on the statistics. The toss-up games are split at 3-3. If Mike had won the 5 games he lost that he deserved to win and lost the 1 game he won that he deserved to lose then his record would be 23-14 for 62% winners. Even if you don’t agree with one of the games on my list he still deserves to be 22-15 for 59% winners, which is right at his lifetime percentage.
The lesson is to not judge a handicapper solely on his short term record, but rather on his long term record and the circumstances of his short term record should be taken into account, as short term results are subject to variance, both positive and negative, while the variance evens out over time. The season Mike is having reminds me a lot of my 2011 College season, in which I suffered a great deal of negative variance with fumbles (-19) and close games (won just 30%) and finished just 1 game above .500 rather than being 58%, which is what I graded myself at that year. The very next season I was 58.5% handicapping at the same level I did in 2011 but without the negative variance. Variance works both ways and there are years in which Mike or I have won at a higher rate than we handicapped at.
I'm not making excuses for Mike's 51% record so far this season but rather just stating the facts that his handicapping has been as good as it always has been. If Mike continues to handicap at the level he has this season, and over his 8 year career as a public handicapper then he will win going forward just as he's won in the past.
Here’s my list with some details on why each game is in the category I put them in.
Undeserved wins (1)
Week 2 Tampa Bay (+3) 14-16 over New Orleans :
Undeserved Losses (5)
Week 4 Houston (+1 ½) 20-23 vs Seattle:
Week 5 Jacksonville (+11 ½) 20-34 vs St. Louis:
Week 6 Green Bay (-2 ½) 19-17 vs Baltimore: The Packers led this game 16-3 in the 4th quarter despite losing 2 of their top 3 receivers to injuries early in the game, leaving Aaron Rodgers with just 2 healthy wide receivers for 3 ½ quarters. Green Bay still outgained Baltimore 440 yards at 7.0 yppl to 360 yards at 5.9 yppl and Baltimore’s spread covering touchdown with just over 2 minutes left in the game was possible because of a conversion on 4th and 21 yards. Green Bay, up by 2, ended the game with 1st and goal at the Baltimore 9 yard line and took two knees to end the game. Brutal loss.
Week 9 Buffalo (+3) 13-23 vs Kansas City: Buffalo outgained Kansas City 470 yards at 6.1 yppl to 213 yards at 4.2 yppl but were -3 in turnovers with two of those turnovers being returned for touchdowns (one for 100 yards). Buffalo was a great bet.
Week 9 Houston (+2) 24-27 vs Indianapolis:
Toss Up Wins (3)
Week 3 Miami (-2 ½) 27-23 over Atlanta: Turnovers were even and Miami had 5.3 yppl to 5.5 yppl for Atlanta.
Week 3 Kanas City (+3) 26-16 over Philadelphia:
Week 6 Buffalo (+7) 24-27 (OT) vs Cincinnati:
Toss Up Losses (3)
Week 5 Tennessee (+2 ½) 17-26 vs Kansas City: The total yards were very even at 340 yards at 5.2 yppl for Tennessee and 353 yards at 5.3 yppl for Kansas City, but the Chiefs scored a touchdown on a fumble and had two 4th quarter interceptions in the final minutes that resulted in made field goals despite not moving the ball on those drives. This game was directly affected by the turnovers and ultimately that touchdown on the fumble was the difference between covering and not.
Week 6 Oakland (+8 ½) 7-24 vs Kansas City:
Week 9 Cincinnati (-3) 20-22 vs Miami:
There have been 5 other games that have been within 7 points of the spread that weren’t really toss up games.
Week 2 Buffalo (+3) 24-23 over Carolina: 436 yards at 6.0 yppl to 308 yards at 4.1 yppl and won despite -1 in turnovers.
Week 3 Jets (-2 ½) 27-20 over Buffalo:
Week 7 Jets (+4) 30-27 over New England:
Week 8 Green Bay (-7) 44-31 over Minnesota:
Week 9 Minnesota (+10) 23-27 vs Dallas:
For the season Mike,s NFL Best Bets are just 19-18 for -6.6 Stars but I'm very confident that he'll have a profitable record by the end of the season given how well he's actually handicapped this season and given his 307-220-13 (58.3%) record on NFL Best Bets in his career as a public handicapper (since 2006).
NFL Free Analysis
As most of you know, Dr Bob has posted informative and profitable NFL analysis in his Free Analysis section for years. This season I will post Free Analysis on NFL games that are not part of my Best Bet package.
When the sample gets large enough, you have to stop saying you've been "unlucky". It may just be that your model is not factoring in all relevant variables.
ReplyDelete