SoSH Survivor Pool - Week 1 Discussion

Stitch01 · Sep 11, 2015

Yeah, calling that a tossup seems crazy. All the evidence says Dallas is the better team and they are playing at home. That's not really how lines work, not to the extent of a game that's a true toss up going out at -6. Survivor grid has it as the 2nd highest EV pick of the weekend, which seems reasonable to me.

Football outsiders has the true line at Dallas -5.6 and Dallas as the 7th most likely team to win this week

glennhoffmania · Sep 11, 2015

FL4WL3SS said:
I don't really care what the lines are, those are meant to entice betting not predict who will win the game. These teams intimately know each other, it's a toss-up game if there ever was one.

I get that divisional games can be tougher but I disagree that lines are meaningless. There are reasons why many people are willing to give 6 points. This isn't a huge deal either way. I like all three games. I just personally don't see Miami (or even GB, which is another divisional game anyway) as clearly better than Dallas at home. So my feeling is that we diversify in case something fluky happens in any of these three games.

wade boggs chicken dinner · Sep 11, 2015

The reason we are picking MIA is because we think the Redskins are going to be terrible.

Is that true for NYG? I'm honestly asking because I follow the AFC more than the NFC.

I'm just seeing Cruz is likely out; if he doesn't play, that would be a significant factor in my thinking.

glennhoffmania · Sep 11, 2015

Cruz is out, as is JPP.

FL4WL3SS · Sep 11, 2015

NM, read that wrong

wade boggs chicken dinner · Sep 11, 2015

Assuming we are picking 4 teams and assuming that GB gets 8 of the remaining picks, I think we probably should do something closer to 50-50 on DAL and MIA (rather than 8 on MIA and 1 on DAL) - not because I think DAL is a better game but if we are going to diversify, we really should diversify. Something like 6 MIA and 3 DAL makes a lot more sense (strategically - not who is going to win) to me than 8 MIA and 1 DAL.

glennhoffmania said:
Cruz is out, as is JPP.

Not that it matters much to the above post and though I understand that he is not likely to play (hasn't practiced since mid-August) Cruz is not out, at least not officially. http://www.giants.com/news-and-blogs/article-1/WR-corps-ready-to-step-up-if-Victor-Cruz-cant-go/4667dab4-06d8-4e57-a527-569fb2e0a8b8.

glennhoffmania · Sep 11, 2015

Hmm. Ok. Yahoo has him as out but ESPN has him as questionable.

FL4WL3SS · Sep 11, 2015

I like a couple of picks on the NYJ.

wade boggs chicken dinner · Sep 11, 2015

FL4WL3SS said:
I like a couple of picks on the NYJ.

I think we've agreed that we're limiting ourselves to four picks, but in case this is a serious option, I'd pass on this game. Unless the Jets can really run the ball on the Browns, this game is going to hinge on which QB makes fewer mistakes (CLE had I think second most interceptions in NFL last year and third ranked pass defense).

I know some people think the Jets are going to turn into a nine-win team this season (which is funny because BUF and MIA are also supposed to be playoff teams this year) but I think on paper CLE and NYJ are pretty evenly matched.

Stitch01 · Sep 11, 2015

I disagree with limiting ourselves to four picks. I thought we were just tabulating votes and going with it?

If that's consensus, that's fine, but I didn't think we agreed on that, I thought we were just doing it the way we did last year and adding votes up and going with it.

I like the Jets because they match up well with the Browns and we burn very little future value. I certainly wouldn't use many slots on it, but to me its a great place to take a little risk this week.

glennhoffmania · Sep 11, 2015

I have no problem taking a shot on the Jets. We're not going to win if we take no chances. Something like 5-5-5-2 for those 4 games would make sense to me, although I know some people don't like Dallas as much as the other two.

bostonbeerbelly · Sep 11, 2015

I think I have 2 votes left based on my original voting was for 20.

If that is the case - I like the jets as well and want my 2 applied there

Jets - 2

Kenny F'ing Powers · Sep 11, 2015

Picking the winner of 1 game is tough. We aren't going to pick the winner of 4 games. We'll essentially be wasting a few picks every week.

DennyDoyle'sBoil · Sep 11, 2015

Kenny F'ing Powers said:
Picking the winner of 1 game is tough. We aren't going to pick the winner of 4 games. We'll essentially be wasting a few picks every week.

But the object is not to win with every pick, it's to have at least one left at the end, and also to spread around the "used" teams for later weeks. Picking four correct is hard. But picking one out of four correct is easier than picking one out of one correct.

If you put 10 picks on team A and 2 picks on team B, and A wins and B loses, then yes, the 2 picks were "wasted." If team A loses, they were not wasted. They saved you to fight another week. You only know in hindsight. If we had a crystal ball, we would only need one share.

I'm sure that there is some pretty complicated game theory that could be applied to this situation that I'm not smart enough to figure out with respect to optimizing multiple picks in a survivor game. I don't think we have figured it out, and even if one of us understood it, the notion we would get consensus on it seems zero. But I feel pretty confident that the answer is not to use all the picks on one game. If that's our strategy, we dramatically over spent. We should have purchased just one share and split the cost of $75 sixty ways.

Hambone · Sep 11, 2015

I thought that was the point of teaming up to reduce risk. I recall getting burned last year by having too many picks on the same team.

Edit - what DDB said

FL4WL3SS · Sep 11, 2015

Bingo.

And using a team like the Jets early in the season gives those picks much greater future value. It was great winning the Patriots picks, but those picks would have a lower future value than the few we use on the Jets.

This is the best opportunity to use the Jets, IMO.

glennhoffmania · Sep 11, 2015

Also without spreads what percentage of games do you think you can pick correctly? Is 2/3 unreasonable? Anything can happen in one game but the odds of all of us totally whiffing on four games seem pretty low.

I was just reading the survivor article on CBS. I'm not endorsing this guy's views but I figured I'd share. His top three picks are Denver, GB and AZ. His high risk/high reward pick is the Jets. And the game that scares him the most? Miami. Here's why (and it really makes no sense so I'm guessing there was an editing issue):

Dolphins at Redskins: Let me be perfectly frank: there is nothing -- absolutely nothing -- on paper here that indicates Miami should beat Washington. In fact, Miami -3.5 looks like the lock of the century. The Redskins defense isn't good enough to contain Ryan Tannehill and his new weapons (Jordan Cameron, Kenny Stills, Davante Parker, Greg Jennings) or the running game with Lamar Miller.

Kirk Cousins might literally be facing a life/death situation with Ndamukong Suh lined up a few feet from his face. I can't tell you how the Redskins will win this game or even contend in this game but I can tell you my Spidey Sense is going bananas here. Something about this game doesn't feel right and I'm not touching it. It's too easy. Take the Fins if you want but don't say I didn't warn you.

FL4WL3SS · Sep 11, 2015

He messed up and meant there is nothing that indicates that Washington should beat Miami and that it seems too easy of a pick, so he's staying away. Fucking dumb.

FL4WL3SS · Sep 11, 2015

I almost feel like Bill Simmons shadowed that article. Which corollary does that fall under?

glennhoffmania · Sep 11, 2015

Yeah I think Miami is easily a top 3 or 4 pick. I admit that I don't know enough about Denver yet. Is their defense going to suck? The offense looks fine so while I'm not suggesting we use them I don't think it's a horrible pick.

wade boggs chicken dinner · Sep 11, 2015

Stitch01 said:
I disagree with limiting ourselves to four picks. I thought we were just tabulating votes and going with it?

If that's consensus, that's fine, but I didn't think we agreed on that, I thought we were just doing it the way we did last year and adding votes up and going with it.

I like the Jets because they match up well with the Browns and we burn very little future value. I certainly wouldn't use many slots on it, but to me its a great place to take a little risk this week.

I meant this week, not every week. It had been discussed before; maybe there's not a consensus.

glennhoffmania said:
Also without spreads what percentage of games do you think you can pick correctly? Is 2/3 unreasonable? Anything can happen in one game but the odds of all of us totally whiffing on four games seem pretty low.

DennyDoyle'sBoil said:
But the object is not to win with every pick, it's to have at least one left at the end, and also to spread around the "used" teams for later weeks. Picking four correct is hard. But picking one out of four correct is easier than picking one out of one correct.

If you put 10 picks on team A and 2 picks on team B, and A wins and B loses, then yes, the 2 picks were "wasted." If team A loses, they were not wasted. They saved you to fight another week. You only know in hindsight. If we had a crystal ball, we would only need one share.

I'm sure that there is some pretty complicated game theory that could be applied to this situation that I'm not smart enough to figure out with respect to optimizing multiple picks in a survivor game. I don't think we have figured it out, and even if one of us understood it, the notion we would get consensus on it seems zero. But I feel pretty confident that the answer is not to use all the picks on one game. If that's our strategy, we dramatically over spent. We should have purchased just one share and split the cost of $75 sixty ways.

Here's how I see the game theory perspective - I'm not a statistician so I'm sure someone will correct me if I'm wrong.

Theoretically, we could assure ourselves of a share of winning by buying enough entries (and using a computer) to put 1/2 of our entries on both sides of the same game. No one does that because it's obviously cost prohibitive.

But to take this to a practical level, it seems to me that the best strategy is to preserve as many picks at a late enough date so as to be able to pick both sides of the same games for the last several weeks to ensure victory. For example, if you had two picks on Week 17 and hadn't used KC and OAK, you could put one on KC and one on OAK in order to ensure one is left.

If this is correct, the ideal way of implementing this is to pick only one game. After all, if a good pick has a 70% chance of winning, trying to pick two games reduces our chance of getting all of our entries through from 70% to 49%. Three games is under 35%.

However, we're also in this to have fun. So no one wants to put 22 picks on any team and see a last-second Josh Scobee FG to tie the game and send us out after one week. So the more picks we make, the higher the odds we get some picks through but the lower the odds that we have all 22 picks.

To me, the takeaway is this - try to get as many picks through the early weeks by making fewer picks and focusing on teams that we think are going to be terrible so hopefully we can have a strategic advantage in the later weeks. I'm not sold on MIA being a good team, but I'm pretty sure the Redskins are going to be horrible (at least I hope they are, if just for one week).

That's just what I've been thinking though.

Stitch01 · Sep 11, 2015

If we have 2 picks that we think are 70% to win in week 1 we almost for sure should be reducing variance and splitting the picks.

I think the biggest strategic advantage you can build in these things later is to have teams left that other people don't have. So if you can find a matchup you like with a team without much future value early and can save one of the likely big guns for later, you will be in better shape in the end game by being able to either have a team much more likely to win or fade the chalk pick.

But yeah, for now we should mostly be trying to get teams through, that can change as we go and in end game the correct play can be the team less likely to win even ignoring future value.

tonyandpals · Sep 11, 2015

wade boggs chicken dinner said:
But to take this to a practical level, it seems to me that the best strategy is to preserve as many picks at a late enough date so as to be able to pick both sides of the same games for the last several weeks to ensure victory. For example, if you had two picks on Week 17 and hadn't used KC and OAK, you could put one on KC and one on OAK in order to ensure one is left.

Just want to mention that you're likely picking 2 teams per week from week from week 14 on. In week 17 you'd need 4 teams to pull that move off. They'd all have to be teams you haven't picked, playing each-other. Seems like that would be pretty tricky to line up logistically.

wade boggs chicken dinner · Sep 11, 2015

Just want to mention that you're likely picking 2 teams per week from week from week 14 on. In week 17 you'd need 4 teams to pull that move off. They'd all have to be teams you haven't picked, playing each-other. Seems like that would be pretty tricky to line up logistically.

Yeah you're right; I forgot about the two picks. I was really just thinking aloud. Seems to me that concentrating picks in the early weeks on early weeks is the best strategy for us but maybe it is the right strategy to use a couple of picks each on teams with very low expected future value - NYJ, TB, and CIN come to mind - but it seems to me that things are volatile enough in the early weeks of the NFL that we should start worrying about future value a bit later in the process.

I certainly admit that I could be wrong and would be willing to be convinced.

glennhoffmania · Sep 11, 2015

Anyone alive in week 14 has to start picking two teams?

tonyandpals · Sep 11, 2015

glennhoffmania said:
Anyone alive in week 14 has to start picking two teams?

From the rules (sorry if I neglected to post, but I think we covered it last year too)

Like the past few seasons, you'll have to pick 2 teams to win for the last 4 weeks if there are over a certain # still alive. I've done this to try to cut down on the possibility of multiple winners at the end of the season... here's how it will work:

If after week #13 we still have over 100 entries alive, you will have to select 2 teams to win in week 14.

If after week #14 we still have over 50 entries alive, you will have to select 2 teams to win in week 15.

And if after week #15 we still have over 25 entries alive, you will have to select 2 teams to win in week 16.

Lastly, if after week #16 we still have over 10 entries alive, you will have to select 2 teams to win in week 17.

fake edit: team count is up to 9200 / 690k

Kenny F'ing Powers · Sep 11, 2015

So, if we pick 4 teams this week, can we never pick those ones again? How are we keeping track of which pick has picked which team?

tonyandpals · Sep 11, 2015

Kenny F'ing Powers said:
So, if we pick 4 teams this week, can we never pick those ones again? How are we keeping track of which pick has picked which team?

For each entry a team can only be selected once. If we pick Pats on Entry1 in week 1, we can't use Pats on Entry1 for the rest of the year. We have 22 entries, all independent of eachother.

I am keeping the master sheet of what teams we've used on which entries and will post it each week.

tonyandpals · Sep 11, 2015

Here is the sheet

tonyandpals · Sep 11, 2015

Here are the %s and my opinion

NE - 5 (locked)
MIA 6.122
GB 6.026
NYJ 1.626
DAL 1.626

I've been listening in and I think we need to settle in on a scenario. I'm just going to put them out there, you guys can bold the one you want and I'll capture it. Avoiding a true "poll" as non-shares may vote and skew. Please bold your selected scenario so it jumps out at me (and everyone else). You can also make up your own and people can vote on it. Locking the vote at 11AM on Sunday AM.

Scenario 1:
MIA(7), GB (6), NYJ (2), DAL (2)
Miami wins the popular vote, earns the extra 1. Any team that registered a 1.5 earns picks.

Scenario 2:
MIA(9), GB (8)
The 3 team approach. Miami and GB earn their share per the vote, we drop the two 1.5s.

Scenario 3:
MIA(7), GB (6), NYJ (4)
We don't touch Dallas

Scenario 4:
MIA(7), GB (6), DAL (4)
In Romo we trust

Scenario 5:

MIA(6), GB (6), NYJ (2), DAL (2), MIN (1)
Spread 'em - MIN was the next highest at .6

Scenario 6:

MIA(5), GB (5), NYJ (3), DAL (3), MIN (1)
Spread 'em thinner

OTHER:
Pitch and bold your scenario. Others can then vote on it if they like.

This would have been a lot easier if we just went 22 on the Champs.

mauidano · Sep 12, 2015

tonyandpals said:
Here are the %s and my opinion

NE - 5 (locked)
MIA 6.122
GB 6.026
NYJ 1.626
DAL 1.626

I've been listening in and I think we need to settle in on a scenario. I'm just going to put them out there, you guys can bold the one you want and I'll capture it. Avoiding a true "poll" as non-shares may vote and skew. Please bold your selected scenario so it jumps out at me (and everyone else). You can also make up your own and people can vote on it. Locking the vote at 11AM on Sunday AM.

Scenario 1:
MIA(7), GB (6), NYJ (2), DAL (2)
Miami wins the popular vote, earns the extra 1. Any team that registered a 1.5 earns picks.

Scenario 2:
MIA(9), GB (8)
The 3 team approach. Miami and GB earn their share per the vote, we drop the two 1.5s.

Scenario 3:
MIA(7), GB (6), NYJ (4)
We don't touch Dallas

Scenario 4:
MIA(7), GB (6), DAL (4)
In Romo we trust

Scenario 5:

MIA(6), GB (6), NYJ (2), DAL (2), MIN (1)
Spread 'em - MIN was the next highest at .6

Scenario 6:

MIA(5), GB (5), NYJ (3), DAL (3), MIN (1)
Spread 'em thinner

OTHER:
Pitch and bold your scenario. Others can then vote on it if they like.

This would have been a lot easier if we just went 22 on the Champs.

Scenario 3:
MIA(7), GB (6), NYJ (4)

We don't touch Dallas

fiskful of dollars · Sep 12, 2015

tonyandpals said:
Here are the %s and my opinion

NE - 5 (locked)
MIA 6.122
GB 6.026
NYJ 1.626
DAL 1.626

I've been listening in and I think we need to settle in on a scenario. I'm just going to put them out there, you guys can bold the one you want and I'll capture it. Avoiding a true "poll" as non-shares may vote and skew. Please bold your selected scenario so it jumps out at me (and everyone else). You can also make up your own and people can vote on it. Locking the vote at 11AM on Sunday AM.

Scenario 1:
MIA(7), GB (6), NYJ (2), DAL (2)
Miami wins the popular vote, earns the extra 1. Any team that registered a 1.5 earns picks.

Scenario 2:
MIA(9), GB (8)
The 3 team approach. Miami and GB earn their share per the vote, we drop the two 1.5s.

Scenario 3:
MIA(7), GB (6), NYJ (4)
We don't touch Dallas

Scenario 4:
MIA(7), GB (6), DAL (4)
In Romo we trust

Scenario 5:

MIA(6), GB (6), NYJ (2), DAL (2), MIN (1)
Spread 'em - MIN was the next highest at .6

Scenario 6:

MIA(5), GB (5), NYJ (3), DAL (3), MIN (1)
Spread 'em thinner

OTHER:
Pitch and bold your scenario. Others can then vote on it if they like.

This would have been a lot easier if we just went 22 on the Champs.

Like scenario #3.

Scenario 3:
MIA(7), GB (6), NYJ (4)
We don't touch Dallas

wade boggs chicken dinner · Sep 12, 2015

Are we really pitting 18% of our picks on a team with a new QB and a new coach? This seems like outsmarting ourselves like last year.

538 mentioned in a preview that - http://fivethirtyeight.com/features/2015-nfl-preview-the-cowboys-eagles-and-giants-look-good-compared-to-the-disaster-in-washington/ - the average team loses 17% of playoff odds when they swap out both coach and QB. Maybe Todd Bowles is going to be a great head coach but he could be Rod Marinelli too.

If we are going to diversify, I'd go with Scenario 1 but I'd say our game theory bet for this week is scenario 2.

FL4WL3SS · Sep 12, 2015

wade boggs chicken dinner said:
Are we really pitting 18% of our picks on a team with a new QB and a new coach? This seems like outsmarting ourselves like last year.

538 mentioned in a preview that - http://fivethirtyeight.com/features/2015-nfl-preview-the-cowboys-eagles-and-giants-look-good-compared-to-the-disaster-in-washington/ - the average team loses 17% of playoff odds when they swap out both coach and QB. Maybe Todd Bowles is going to be a great head coach but he could be Rod Marinelli too.

If we are going to diversify, I'd go with Scenario 1 but I'd say our game theory bet for this week is scenario 2.

Well we already have five picks in the bank.

[bold]Scenario 3[/bold]

glennhoffmania · Sep 12, 2015

Scenario 1:
MIA(7), GB (6), NYJ (2), DAL (2)

Sorry don't see how to bold on phone.

Stitch01 · Sep 12, 2015

Ha same with the bold issue

SCENARIO ONE

Prefer five but that has no support

j44thor · Sep 12, 2015

FL4WL3SS said:
Well we already have five picks in the bank.
[bold]Scenario 3[/bold]

The average team does not add Revis, Cromartie and Skrine to a loaded defense.
Change at QB should also be a net positive.

I like scenario three.

glennhoffmania · Sep 12, 2015

Cruz has been ruled out.

Papelbon's Poutine · Sep 12, 2015

scenario 2

bostonbeerbelly · Sep 12, 2015

Scenario 1:
MIA(7), GB (6), NYJ (2), DAL (2)

Ed Hillel · Sep 12, 2015

Stitch01 said:
Ha same with the bold issue

SCENARIO ONE

Prefer five but that has no support

I'm with you on Scenario 5.

I'd take the Vikings over Jets in a heartbeat, but that's just me I think. That Jets offense could score less than 10 points against Cleveland.

wade boggs chicken dinner · Sep 12, 2015

I'm with you on Scenario 5.

I'd take the Vikings over Jets in a heartbeat, but that's just me I think. That Jets offense could score less than 10 points against Cleveland.

If there was enough support for Scenario 5, I'd vote for it. Basically, I would very much prefer not to put 4 shares (18%) on the Jets.

CLE was 6-3 at one point last year; they seem to have upgraded their running defense with Danny Shelton; they have a top-five passing defense; and they get Alex Mack back. When Alex Mack was in playing last year they had a top-10 rushing attack. He's a big key to their team. Plus the O/U is 40 (the lowest of the week) and I suspect that the under is the better play.

If we were to rate games like we do fantasy football players, I would put NE / GB / MIA in one tier and then DAL / CIN / MIN in a second tier. I would actually put NYJ in a third tier with SEA + PHI + CAR. But to the extent we are going to diversify past the first tier, I'd like the picks spread out as much as possible.

chief1 · Sep 12, 2015

Im in the camp of, we need to take chances to win this thing. We will have used up two of the three best teams (NE & GB) on a lot of cards the first week of the season and it will be good strategy to get past week one (or any week) with weaker teams. I think the NYJ will actually be competitive this year and think Cleveland is still way to dysfunctional I also think SF and Wash will be choosing in the top 5 of the draft next year.

I am choosing Option 5.

SoSH Survivor Pool - Week 1 Discussion

Member

meat puppet

Member

meat puppet

Member

Member

meat puppet

PN23's replacement

Member

PN23's replacement

Member

Member

meat puppet

Member

posts way less than 18% useful shit

Found no thrill on Blueberry Hill

will post for drinks

Member

PN23's replacement

meat puppet

Member

PN23's replacement

Member

meat puppet

PN23's replacement

Member

Member

Well-Known Member

Member

PN23's replacement

meat puppet

Well-Known Member

posts way less than 18% useful shit

Well-Known Member

Well-Known Member

Well-Known Member

Mai Tais for everyone!

Well-Known Member

PN23's replacement

Member

Member

meat puppet

Member

Member

meat puppet

Homeland Security

Member

Wants to be startin somethin

Member

New Member