Wednesday, October 23, 2013

What Was Optimal? Looking Back on 10 Weeks of C.H.A.D. Picks

I thought it'd be fun to take a look back at how the weekly C.H.A.D. picks look when compared using the Pinny closing ML and how popular each team was. So, this is that.

Worth noting that because of the double-elimination format of the preseason, the standard SR/popularity EV calculation doesn't really hold. So I definitely did some #makingshitup to attempt to account for this, which definitely isn't perfect, but #whocares.

At least I removed the vig from the Pinny lines? Anyway, here goes, and please stick around until (or just scroll straight to) the bottom for a possible C.H.A.D. 2014 idea.

 

The Colts ended up being the best pick here (that was me!), but I think that was mostly because they got bet a bunch after the pick submission deadline. The Browns' absurd popularity hurts them, and the Jags were an underdog so that probably wasn't advisable, but at least @alexrolfe wasn't giving up much future value!



Not really sure what @uug9's thought process behind picking Denver was here, but it did not work out. Despite Chicago's popularity, they don't end up looking so bad under this methodology since they were favored by more than anyone other than the New Orleans. Speaking of the Saints, other than future value I'm no sure why they were so unpopular here, but regardless it worked out for @matt_matros.


The Denver pick (also made by @matt_matros) comes out looking pretty good here, but obviously that's ignoring the substantial FV sacrifice. Some team has to be the most popular of course, but given the second- and third-leveling going on in this contest, I think it's funny/interesting that a team was >40% for each of the first three weeks of preseason.

X4 was weird, as a lot of people didn't care about their result since they had two strikes remaining, so let's move on to the regular season. Single-elimination now, so popularity is back to being factored in the standard way.


That's an ugly EV for Pittsburgh. They ended up being much more popular in C.H.A.D. than in OFP (11.5%), so that was part of it. And then that line got bet down on Sunday, and Pouncey's knee blew up on the first possession of the game, and it was all downhill from there for over a third of the pool.

The Pats' OFP popularity (17.6%) didn't quite translate to C.H.A.D., so New England ended up being decent even after you dock them for future value. The same cannot be said for @harmyg's Seattle pick.

All over the place. It wasn't so obvious at the time, but the Texans ended up being pretty good due to their lack of future value. Both Chicago and Green Bay were a lot more popular in C.H.A.D. than OFP, and a second person taking Atlanta really killed their EV.


All four of these teams had quite a bit of future value; I suspect the three folks who took New England didn't expect them to be so popular, as the Pats were only picked on 2.2% of OFP entries. Seattle probably comes out looking the best here.


Looks like Indy was the best pick here. Kansas City was not so great, in another situation where a second person making a seemingly obscure pick ruins it for both people.


Whether due to luck or good planning, this really worked out for the two people who had yet to use St. Louis.


Another situation where everybody who could possibly pick against the Jags did just that, and didn't regret it. The rest of this shakes out pretty reasonably after you consider FV, with Chicago probably ending up as the second best pick.


Impressive. The Packers, Dolphins, and Falcons all come out looking pretty good here.

Here are the picks and EVs for our six remaining entries, again excluding X4:


And finally, the EV standings, as it were:


With only one "bad" pick (ARI in X3), @notthattricky comes out on top here. Kind of impressive that he's ahead of @cajuncooks, who was able to make very strong picks in both W5 (STL) and W6 (DEN).

Not to mess with perfection, but for C.H.A.D. 2014 I think it might be fun for half of it to be the same as this year's C.H.A.D., and the other half to be based solely on EV according to Pinny close, ignoring the actual results of the game. This would have the added benefit of allowing everybody to make picks for all 17 weeks, even after being eliminated from the more traditional half of the pool.

2 comments:

  1. It's possible that the X4 issue could be solved by either not having preseason be double elimination, or by not having re-buys.

    ReplyDelete
    Replies
    1. I'm not sure there's really an issue, other than for writing a post like this. I didn't make this clear at all, but I was thinking the EV half of the pool next year would exclude preseason.

      One thing we could do with preseason is make it single-elimination, and also reduce the initial buy-in. So more re-buys, but they'd be cheaper.

      Delete