Performify's Pigskin Picks - Week Ten

[Performify]

Gotta love yet another brilliant week last week. Chicago turns the ball over 7 times. Atlanta turns the ball over three times. Green Bay throws an INT in the endzone in the 4th quarter as they're ready to tie up the game, which leads to a Bills TD so 14 point swing. Chargers are covering and give up a garbage TD with a minute left. Yay, double digit losses on the year now. Didn't need the teaser points but nice to see that win without any real danger.

Despite all that, I still have confidence in my system of game selection and weighting. I'm getting burned by turnovers this year: my system is selecting teams which are good at forcing turnovers playing teams that are bad at turning the ball over, and instead in those games my picks are turning the ball over unusually. For example, Chicago was +11 in turnovers (meaning they had generated eleven more turnovers than they had turned the ball over themselves), tied for #1 in the league in that category on the year - at least prior to last week - and Miami was -5. So on average you wouldn't expect that Chicago would give the ball away 6 times for a -4 turnover differential. Likewise with Atlanta/Detroit, where Atlanta is one of the best teams in the league for turnovers playing one of the teams near the bottom in turning the ball over. And instead, Atlanta gives the ball away three times for a -2 turnover differential, and Detroit scores three of their touchdowns off the Atlanta giveaways.

Anyway, I still have all the confidence in my selection system and my capping ability. I really feel this is all short term variance, and I really do expect stronger showings going forward. But feel free to fade away, if you think otherwise [img]/images/graemlins/smile.gif[/img]

Remember, as always on these picks, i'm using graduated units (units are not 1%-2% of my bankroll). If you're playing along at home and are using 1%-2% units, cut my unit size at least in half (6u -> 3u, 1u -> .5u, etc).

On to the picks and writeups, but first: up-to-date record:

Last week:
1-4, -7.24u on straight picks. Loss on Game of the Week.
Win on Tease of the Week, +3u
Total on the day: -4.24u

YTD RESULTS:
13-22-4 on posted picks, -13.63 units (excludes teasers)
2-3-3 on Game of the Week, -10.25 units.
4-3 on Tease of the Week, +1.6 units
Year to date combined on all posted NFL picks (includes teasers): -12.03 units

Jets +10.5 (-107) at Patriots

The Jets are coming off their bye week following a disappointing loss to Cleveland in Week 8. The Pats are coming off a crushing loss to Indy, a game which they gave away with turnovers and penalties.

I look at a lot of factors for these games, including a lot of statistics and breakdowns. With some of the tremendous new statistical sites available, there are a lot of very interesting bits of information available. For one, in 45 career starts at quarterback, Jets quarterback Chad Pennington is 6-2 after games in which he had a quarterback rating lower than 75. Most recently, Pennington had a 28.9 rating against Jacksonville in Week 5 and then torched Miami in Week 6. He had a 21.1 rating against Cleveland in their last game... and now I look for him to continue this trend with a rebound performance against the Patriots. No, before someone jumps on me, not just because of this trend. Its all about the statistics: the Pats give up 229 yards per game in the air on average, making them the #28 ranked pass defense in the NFL. They're coming off a game in which they gave up 326 passing yards in Week 9 vs. the Colts. Also the New England defense does not blitz often, instead preferring to stay in their base 3-4 formation and play the run stronger. And of course, Jets coach Eric Mangini, the Patriots' defensive coordinator last year, knows New England's personnel well and has had two weeks to prepare, so expect some new developments in the Jets' game plan. Patriots starting safety Rodney Harrison will also miss this game, out with a broken shoulder blade, which leaves a big hole in the already leaky New England secondary.

On offense, the Patriots like to run the ball (128 yards per game on the ground on average), and should have a big day against a weak Jets rush defense that allows 148 yards rushing per game. But this is actually a factor in why my system likes this game: The Jets have a high-powered passing attack playing a leaky pass defense, which should allow them to put up points quickly. On the other side of the ball, the Pats will likely pound the run much of the game, which should chew up the clock and serve to keep things closer than most people expect here. Bill Belichick is one of the best coaches in exploiting an opponent's weakness and he will draw up a run-first approach to grind out yards against an undersized Jets defensive line. Another bit of interesting statistical matchup from another of those aforementioned sources: In 11 career games against the Jets, New England QB Tom Brady averages just 196 yards and one touchdown. Worse, in his last three home games against the Jets, Brady has thrown just one touchdown pass.

This will be the second time these divisional rivals have met this year. In week two the Pats won 24-17 in New York. I predict this game will be pretty similar to that last meeting: a game in which the Pats ran for 157 yards, the Jets threw for over 300, and the Pats shut down a late Jets drive to win a close game. 21-14 New England for the Jets cover. Game of the Week.

Chiefs -1 (-103) at Miami

Road favorites can always be trouble, but the Chiefs are on an absolute tear, led by Larry Johnson who has 459 rushing yards out of his last three games. The Dolpins are coming off a big win and I believe there's a chance for an intangible "letdown" game after the big victory over Chicago.

The Dolphins are #14 in the NFL at defending the run, and with the Chiefs offensive line starting to gel and Chiefs backup QB Damon Huard really starting to open up the entire offense, the Chiefs should roll over the Dolphins here. The Chiefs do have a couple struggles at offensive line, with last week's injury to guard Brian Watters, but they have veteran Chris Bober to replace him and will still be able to run against Miami with Bober standing in for Waters.

Joey Harrington will face another Cover 2 defense with the Chiefs, and he's struggled against the Cover 2 throughout his career including last week against Chicago where despite throwing three touchdowns he threw two INTs and made numerous questionable throws. The Chiefs have built a very strong, very underrated defense, and CBs Ty Law and Patrick Surtain will probably each produce an interception against Harrington this game. The Chiefs have allowed just one running back to top 100 yards on the ground this year, and despite Ronnie Brown's turnaround against Chicago I expect him to have a less productive game here, with Miami falling behind early and being forced away from the run. Chicago turned the ball over six times last week, and the Chiefs will protect the ball better than that here. The Dolphins offensive line is slower than average, and the Chiefs defensive front is built around exceptional speed. Look for a big day from the KC defensive ends, especially speedster Jared Allen.

If the Chiefs can limit mistakes, they will win this game. 21-17 Kansas City.

Saints +4 (-102) at Steelers

This is a popular pick this week among some cappers I respect, and my system likes it a little as well. The Steelers are the worst team in the NFL at turning the ball over, at -11 on the year. The Saints aren't great in that category, sitting at -3 on the year, but still I expect the Steelers will not reverse themselves here and play mistake-free football. The Steelers have struggled protecting Roethlisburger all season, and the Saints match up well defensively with an aggressive pass rush. The Saints have allowed only one quarterback to top 250 yards this season, and I do not expect Big Ben to join that elusive club. The Saints defense has also only allowed two running back touchdowns all season, which should mean a rough day for Willie Parker.

On the other side of the ball, the Steelers have allowed some big games for opposing QBs this year, and the combination of Colston and Brees should be able to capitalize to keep things close. The Dynamic Denver Duo of Jake Plummer and Javon Walker tore up the Pittsburgh secondary last week, and you’ll see Brees making big throws between big target Marques Colston and speedster Devery Henderson. Unlike some others I don't think the Saints win here often enough to be worth a moneyline bet, but I do like taking the four points. 24-21 Pittsburgh for the Saints cover.

Tease of the Week: Denver -9.5, Carolina -9.5 2-team, 7-point teaser

I have both of these games being slightly closer than their large spreads, and as such have a tremendous value in moving both inside the three. Denver goes on the road to face an Oakland team that's been playing well lately. They won't play well enough to upset Denver's roll. Carolina faces Tampa Bay at home on Monday night. I think there is greater probability that Carolina doesn't need the points from the tease, but this should be high probability moved inside the three. This is worth a play to win three units in my system.



<ul type="square">
Summary of Picks for the Week[*]Jets +10.5 (-107) at Patriots : 6.42u to win 6u[*]Chiefs -1 (-103) at Dolphins : 3.09u to win 3u[*]Saints +4 (-102) at Pittsburgh : 1.02u to win 1u[*]Teaser: Denver -2.5, Carolina -2.5 : 3.9u to win 3u[/list]
I'll probably be back tomorrow with some more lower-level picks, my system spit out these two games and the tease as pretty strong plays and I'll take a closer look at my factors for the closer 1-unit games.

[MacGuyV]

Kinda like the Jets pick - they usually play close games in Foxboro.
Indifferent on KC - Miami is in the top 5 in average YPC run defense at 3.4; they're #14 in yards because they're always behind.

[Jibba]

I mentioned the Jets pick in my Early Thoughts thread. Even before Rodney went down, I thought 10.5 was a lot considering that it's a divisional game between teams that usually play pretty close games. After watching the Pats without Harrison this past week, I think Pennington will have a good day. So the number seems very high to me and I'm a huge Pats fan. If I had the guts to bet against my team I'd probably be all over it. As it is, I teased NE down to 6 early in the week (with CHI +8.5) and I'll lay off otherwise.

Love the KC pick. They have looked great lately and LJ is going to rip Miami apart. Completely agree with your analysis.

As for the Saints, I went the other way on that one. I just don't buy into the idea that Pitt is done. Eventually we're going to see the real Steelers and I think this will be the week.

Also thinking of following you on the tease but not sure I can still get that Carolina line.

[Performify]

[ QUOTE ]


Also thinking of following you on the tease but not sure I can still get that Carolina line.

[/ QUOTE ]

Its -9.5 -110 on Pinnacle right now

[niss]

Curious why you would not tease the Bears to +8, assuming the spread is still Bears +1.

[dixieflatline]

[ QUOTE ]

Despite all that, I still have confidence in my system of game selection and weighting. I'm getting burned by turnovers this year: my system is selecting teams which are good at forcing turnovers playing teams that are bad at turning the ball over, and instead in those games my picks are turning the ball over unusually. For example, Chicago was +11 in turnovers (meaning they had generated eleven more turnovers than they had turned the ball over themselves), tied for #1 in the league in that category on the year - at least prior to last week - and Miami was -5. So on average you wouldn't expect that Chicago would give the ball away 6 times for a -4 turnover differential.

[/ QUOTE ]

Hey Performify,

I was reading through this thread and came upon this statement and now I have a question about your system. It seems like you rely heavily on turnover differential which is exposing you to the randomness of fumble recoveries.

The bears, for example, came into last week with that gaudy +11 turnover differential but that was almost entirely due to their fumble recoveries. Their offense had fumbled 10 times and lost only 4 (near league average BTW), but their defense had caused 16 fumbles and recovered 14 of them. Clearly Chicago is a very good team, but their luck at recovering fumbles was inflating that.

I completely agree with you, in general, that teams that are creating turnovers and protecting the ball are being underrated by the public. I think if you adjusted your system to using interceptions + total fumbles you might see better results though.

[Performify]

[ QUOTE ]
Curious why you would not tease the Bears to +8, assuming the spread is still Bears +1.

[/ QUOTE ]

I actually have that in an unposted play. Was probably going to come back tomorrow and recommend it.

[Performify]

[ QUOTE ]

Hey Performify,

I was reading through this thread and came upon this statement and now I have a question about your system. It seems like you rely heavily on turnover differential which is exposing you to the randomness of fumble recoveries.

The bears, for example, came into last week with that gaudy +11 turnover differential but that was almost entirely due to their fumble recoveries. Their offense had fumbled 10 times and lost only 4 (near league average BTW), but their defense had caused 16 fumbles and recovered 14 of them. Clearly Chicago is a very good team, but their luck at recovering fumbles was inflating that.

I completely agree with you, in general, that teams that are creating turnovers and protecting the ball are being underrated by the public. I think if you adjusted your system to using interceptions + total fumbles you might see better results though.

[/ QUOTE ]

First off, thanks a bunch for the feedback, and for taking the time to read and think about what I'm doing here.

I actually already use total fumbles in my system, but I do also weight to a slight degree fumbles lost. For the most part, this was built in to penalize the type of fumbles that traditionally occur - for example a team that fumbles a lot with their receivers or special teams almost always lose the fumbles, whereas quarterback or RB fumbles should penalize a little less as they're usually an approximate coinflip to be recovered by either team.

Its actually a very interesting point. I'm going to do some work breaking apart my model and pulling out the lost fumbles portion, and see how much that changes things. I'm also becoming concerned that my model is overrating turnovers entirely. They're not the single major feature of my model, by any means, but they are part of the weighting and given that the last few weeks I've been losing these games with high turnover discrepancies with a high frequency, its giving me cause to consider a tweak to my model.

[checkmate36]

My brother will be at Pittsburg for the NO game so I will be taking NO and the points in an attempt to get him worked up. [img]/images/graemlins/grin.gif[/img] (I planned on taking NO anyways but now it looks like Im trying to give him and his fav team a shot)

[Thremp]

Performify,

Perhaps making several models (tweaking the items you wish to tweak) and running the numbers through and finding how that would change you picks over some random season may make your feel better about just getting bad beat into the ground.