SoSH Survivor Pool - Week 2 Discussion

[glennhoffmania]

 
Remember though, we have to beat like a zillion people to win the pool so we need to be looking at places to bet against the house rather than trying to avoid it.  We're going to have to hit long shots somewhere.  The reward of not picking New Orleans and cutting the pool in half is much better than the reward of picking New Orleans and advancing along with the herd and we can still pick the correct games elsewhere and advance. Getting through when everyone gets through doesnt get us that much, like we had a good week last week but so did anyone else.  So when you say "betting on NO is a losing proposition", that's almost certainly correct if you are talking about it being a 50/50 shot but we dont need NO to lose even close to half the time to make not picking them correct.  The goal isnt to get as many teams as possible through each week, its to have the last team standing.
 
Now all that said given their likely lack of future value and my personal lack of confidence in many of the other semi-large favorites this week, I still want to pick mostly NO but I might be making a mathematical error by doing so.  If this was like TB @ GB, this weekend Id be inclined to underweight heavily, but NO's future value is just likely so limited.

 
The advantage of betting against the house here is also that we can use NO later while most people couldn't.  I know it's a long way off but week 16 would be a great week, at least based on what we know right now, to have NO available for a bunch of picks.  Maybe that's optimistic but if we don't get to week 16 we ain't winning anyway.  So if NO stumbles this week, we don't lose many entries but a lot of people are out.
 
Anyway my comment about the voting was directed at the distribution based on people's bolded picks.  Why don't we first get a total of the percentages and see what that looks like before creating scenarios to vote on? 

 

[tims4wins]

Picking 5 teams seems like a terrible idea this week. We are most likely to go 3-2, perhaps 4-1 at best, unnecessarily siphoning off picks in a week that the majority gets through on NO. This is exactly the type of play I'd be hoping from others if I was going NO heavy this week.

We are basically betting on NO losing this week to gain an edge which is a losing proposition IMO. No need to bet against the house if you don't have to. There will be plenty of tougher weeks to get through, week two is not the time to start out thinking ourselves.

 
While it is true that if we pick 5 teams we are likely to lose at least 1, it also mitigates against a disaster scenario. That being said you do have to take some risks to win this thing.
 
I guess if we use 10 picks on New Orleans and they lose we are no worse off compared to the rest of the pool, whereas if we bet big on Miami and they lose then we lose compared to the rest of the pool.
 
Of course, if we underplay NO and they lose then we benefit big time.

 

[wade boggs chicken dinner]

So we have about 24 hours to decide and I don't think we've figured out a consensus.  Maybe we should punt this week and just distribute our picks with the general population so we will be no better but no worse after this week?
 
The other thing we can do that's kind of a compromise is:  NO = 7; MIA = 7; BAL = 4; IND= 2; PIT = 1.
 
I'm not a huge fan of IND or PIT but I'm including them because they have gotten some votes.
 
Also, we still need to figure out the distribution of any picks.

 

[Kenny F'ing Powers]

So we have about 24 hours to decide and I don't think we've figured out a consensus.  Maybe we should punt this week and just distribute our picks with the general population so we will be no better but no worse after this week?
 
The other thing we can do that's kind of a compromise is:  NO = 7; MIA = 7; BAL = 4; IND= 2; PIT = 1.
 
I'm not a huge fan of IND or PIT but I'm including them because they have gotten some votes.
 
Also, we still need to figure out the distribution of any picks.

I'm confused.

I thought before week 1, we decided to all vote and then place our votes according to the overall numbers. 1 week later, and we're talking about "punting" and simply voting how everyone else in the pool is voting.

Why have we changed our process? Can't we just tally the votes and make our picks accordingly?

 

[Kenny F'ing Powers]

Also, while citing Vegas lines and professional gamblers has it's value, I'd wager almost every person in this tourney is looking at the same stuff. Let's not overuse those pieces.

 

[tims4wins]

I am still in favor of 10-12 on Miami, 5-6 on NO, and the remainder however everyone sees fit - be that Pitt, Indy, Balt, or some combo.

 

[Kenny F'ing Powers]

My vote:

8 - Baltimore
8 - Miami
5 - New Orleans

 

[glennhoffmania]

I am still in favor of 10-12 on Miami, 5-6 on NO, and the remainder however everyone sees fit - be that Pitt, Indy, Balt, or some combo.

Same here except that I put all 5 extras on BAL.

 

[JoePoulson]

So I'm in another pool with 3688 entries left.  Here's the pick breakdown for those interested, and since we're seemingly at an impasse:
 
NO - 1406 / 38.1%
MIA - 638 / 17.3%
BAL - 582 / 15.8%
IND - 362 / 9.8%
PIT - 241 / 6.5%
STL - 196 / 5.3%      <--- Dang
 
Nothing earth-shattering here, just some more data for us to evaluate.

 

[FL4WL3SS]

So Tony hasn't been in this thread since Thursday, what happens if he misses putting in our picks?

 

[tims4wins]

He will be around, and we have a backup login

 

[Buffalo Head]

My vote:
 
MIA: 6
NO: 6
BAL: 5
PIT: 4

 

[mauidano]

My vote:
 
MIA: 6
NO: 6
BAL: 5
PIT: 4

 
I'm comfortable with these teams and a reasonably close allocation of the picks.  Some of the weeks area easier than others.  Too much hand wringing.  There will be upsets along the way, part of the game.  Put 'em in and let's see what happens.  Good luck to us!!!!

 

[glennhoffmania]

Has anyone totalled up the votes so we can get an idea of where we're at?

 

[tims4wins]

Not to my knowledge. Would be good to know. Maybe we just go with that since we haven't come to a consensus scenario.

 

[AirborneSas]

MIA: 6
NO: 6
BAL: 5
PIT: 4

 

[bostonbeerbelly]

I tried my best to add up the votes cast taking into account when people changed their votes. Easily could have missed one or two, so if someone wants to correct this please do.

MIA - 96 votes 28.8%
BAL - 59 votes 17.7%
NO - 122 votes 36.6%
PIT - 27 votes 8.1%
STL - 3 votes .9%
ARI - 8 votes 2.4%

 

[tims4wins]

That would equate to
NO 8
Mia 6
Bal 4
Pit 2

Are there no Indy votes?

 

[bostonbeerbelly]

On mobile sorry
IND - 18 votes 5.4%

I could be off by a few votes here or there but gives us the general consensus.

 

[bostonbeerbelly]

NO - 8
MIA - 6
BAL - 4
PIT - 2
IND - 1

 

[tims4wins]

NO - 8
MIA - 6
BAL - 4
PIT - 2
IND - 1

 
Thanks, and thanks for compiling. I think we should go with this unless there are major objections.
 
Edit: and I would put 6 of the 8 NO entries on the Miami week 1 entries, and the remaining 2 on the week 1 Jets entries... not sure how to distribute the rest. Maybe 3 of the Miami ones on the Pats week 1 and 3 on GB week 1... then 2 Balt on Pats and 2 Balt on GB... Then 1 Pitt on GB and 1 Pitt on Dallas... then the Indy vote on the Dallas week 1 entry

 

[bostonbeerbelly]

It seems like a very conservative approach and I am okay with that for week 2. I expect us to be in the same position relative to the field when the week is over.

 

[glennhoffmania]

I really think that the Indy pick should be further discussed.  They have a lot of potential future value and this week scares the crap out of me, especially with Hilton possibly out.

 

[tims4wins]

I am fine with scrapping it and putting it on NO or Miami since putting 1 share on Indy seems pointless. If they lose then we gain relative to the rest of the pool and even if they win we save them

 

[glennhoffmania]

Agreed.

 

[wade boggs chicken dinner]

NO - 8
MIA - 6
BAL - 4
PIT - 2
IND - 1

If we are scrapping the IND pick, can we put it on BAL? That would be:

NO = 8
MIA = 6
BAL = 5
PIT = 2

I would be good with this.  Agreed that it's pretty conservative and won't move the needle one way or the other.

 

[tims4wins]

I am fine with that too. Doesn't really matter to me but no point in one pick on Indy

 

[JoePoulson]

Yea that works for me as well.

 

[tonyandpals]

Just got in from running a fundrasier all day. I had the totlas through KFP the other day. I'll add the rest now.
 
edit; Not through KFP, because I worked backwards through the thread and he was fist. Through spaulding I believe.

 

[tonyandpals]

So Tony hasn't been in this thread since Thursday, what happens if he misses putting in our picks?

Not going to happen. I think I mentioned things were busy this week. Event is over and I am back. There is a backup w/ the credentials too.

 

[tonyandpals]

My counts:
 
Just 6 teams w/ votes, one of which had 1.
 
BAL 3.6
IND 1.2
MIA 5.7
NO 8.4
PIT 1.7
STL 0.1
 
I'm leaning to just 3 games
10 / 6 / 5

 

[crystalline]

I tried my best to add up the votes cast taking into account when people changed their votes. Easily could have missed one or two, so if someone wants to correct this please do.

MIA - 96 votes 28.8%
BAL - 59 votes 17.7%
NO - 122 votes 36.6%
PIT - 27 votes 8.1%
STL - 3 votes .9%
ARI - 8 votes 2.4%

I don't have any shares, but about process:
If you use this method of tabulating votes and assigning picks each week based on total votes gotten, you're very likely for the picks each week to follow the overall pick percentage in the pool. That's because the mean over a lot of people here with different opinions is likely to be near the mean over the whole pool. Said another way: one person here might want NO overrepresented so would vote heavily for them. Another person would want them underrepresented so assigns few votes. Using this system neither person gets what they want and instead you get a mix that's very close to the allocation of the whole pool.

Bbb's tallies above show this effect-- the vote percentages posted above are close to the overall pool percentages.

This week I think that's OK, picking the same as the pool makes sense. But going forward you may want to use the multiple-scenario voting system to allow the possibility of over picking or under picking teams relative to the overall pool.

 

[tims4wins]

Crystalline, I think we all totally understand that. At some point we need to go against the grain. But it doesn't seem like we have consensus to do so this week

 

[crystalline]

Crystalline, I think we all totally understand that. At some point we need to go against the grain. But it doesn't seem like we have consensus to do so this week

OK, apologies if that was obvious...

 

[bostonbeerbelly]

I am fine with any of the scenarios above. I guess my instinct is let's keep pace with the field for the first few weeks and when we have more data we can go against the grain when we see an opening. We have 21 entries we can take some chances down the road if we keep a lot alive. Many people have 1 or 2 entries and can be done in one upset, spreading our risk early on makes sense to me instead of going for the home run pick in week 2.

 

[wade boggs chicken dinner]

I don't have any shares, but about process:
If you use this method of tabulating votes and assigning picks each week based on total votes gotten, you're very likely for the picks each week to follow the overall pick percentage in the pool. That's because the mean over a lot of people here with different opinions is likely to be near the mean over the whole pool. Said another way: one person here might want NO overrepresented so would vote heavily for them. Another person would want them underrepresented so assigns few votes. Using this system neither person gets what they want and instead you get a mix that's very close to the allocation of the whole pool.

 
Right - we had talked about that earlier in the thread and more last week, and that's why I thought we had agreed to have a preliminary tally but our final decision would be based on "strategies" or "scenarios." Which is why - to answer KFP - I tried to put together some scenarios and one final scenario I thought might be a consensus pick.
 
It still seems to me that there must be some way to leverage our collective wisdom, but I'm kind of at a loss on how to do it.

 

[crystalline]

 
 
It still seems to me that there must be some way to leverage our collective wisdom, but I'm kind of at a loss on how to do it.

Yeah. If you're going to have a bunch of people vote for each entry's choice and take the average, you almost might as well just use the total pool percentages and save a bunch of work tabulating votes.

It would be interesting, and fairly simple, to use past years' results and figure out how many entries you'd need to win if you just mirrored the overall pool percentages. Happy to advise if anyone wants to tackle that, but too lazy to do it myself.

Enjoy watching the games, all.

 

[Cumberland Blues]

I haven't had time to play with the different scenarios as I'd hoped - but maybe I still can later tonight.  Have others thought on how to divy up the picks?  Some decisions are likely dictated by picking MIA in both weeks - obviously the week 2 MIA picks have to be split among the non-MIA picks from week 1.  My gut instinct is to spread things around as much as possible - thinking that opens up the most combinations for later - but I'm not certain that instinct is correct.  Just thinking out loud a bit....if we have 8 NO picks....do we go say, 2 each on MIA/GB/NE tickets and one each on DAL/NYJ?  Or do we load 'em all up on just two combos - say 5 NE and 3 MIA...I dunno - seems there should be an optimal solution to how to divide them up, but it gets complicated pretty quick when you start taking into account the complexities of the schedule.

 

[Cumberland Blues]

More thoughts on how to spread things around - it seems like a lot of you have a rough idea of which teams carry the most future value...does it make sense to pair the week 2 picks w/ low future value with the higher fv picks we used up in week1? 

 

[tims4wins]

I wrote about it earlier but I think the best approach is to use the NO entries on the Miami and Jets entries from week 1 - that way we would have a TON of future value left on those entries as opposed to putting some on GB or NE

 

[Kenny F'ing Powers]

 
Right - we had talked about that earlier in the thread and more last week, and that's why I thought we had agreed to have a preliminary tally but our final decision would be based on "strategies" or "scenarios." Which is why - to answer KFP - I tried to put together some scenarios and one final scenario I thought might be a consensus pick.
 
It still seems to me that there must be some way to leverage our collective wisdom, but I'm kind of at a loss on how to do it.

We need to watch a lot of football and hopefully we'll have enough matchups knowledge to cut against the grain ocassionally.

 

[mauf]

Also, while citing Vegas lines and professional gamblers has it's value, I'd wager almost every person in this tourney is looking at the same stuff. Let's not overuse those pieces.

I'll bet we could count on one hand the number of people in this thread who can pick the winner ATS 55% or more of the time. We're not going to win because we know football better than the masses. We might, however, be able to solve for the unique problem before us -- maximizing the chance of winning a massive survivor pool with 20ish entries -- better than the other people in that pool.

I'd much rather have people cite Vegas odds as Gospel truth and use them to brainstorm winning pool strategies than have those people go on and on about how they like one 2-1 favorite and not another, based on no-shit factors that are clearly baked into the odds. Those are the posts that don't add value, imo.

 

[Kenny F'ing Powers]

I'll bet we could count on one hand the number of people in this thread who can pick the winner ATS 55% or more of the time. We're not going to win because we know football better than the masses. We might, however, be able to solve for the unique problem before us -- maximizing the chance of winning a massive survivor pool with 20ish entries -- better than the other people in that pool.

I'd much rather have people cite Vegas odds as Gospel truth and use them to brainstorm winning pool strategies than have those people go on and on about how they like one 2-1 favorite and not another, based on no-shit factors that are clearly baked into the odds. Those are the posts that don't add value, imo.

ATS doesn't mean anything here.

 

[mauf]

ATS doesn't mean anything here.

As you know, ATS is usually a good proxy for the money line. Point is, almost no one here can fix the probabilities better than the Vegas lines.

 

[tonyandpals]

guys i'm totally drained so I'll be checking in tomorrow morning and my proposal is we go
 
9/6/4/2 based on the %s this week
 
Kick around anything you like and make your case to others and when I wake up tomorrow and settle in to make the picks around 11, I'm hoping to see some sort of scenario has one out.

 

[glennhoffmania]

I wrote about it earlier but I think the best approach is to use the NO entries on the Miami and Jets entries from week 1 - that way we would have a TON of future value left on those entries as opposed to putting some on GB or NE

Yeah I've been trying to articulate this all week. I think for as many as the week 1 Miami picks as possible we do NO in week 2. Assuming NO wins we'll have a bunch of entries that get through using two of the weaker teams, and NE, BAL, PIT, and GB would all be available going forward. I see no reason to spread it around and weaken some entries.

 

[chief1]

Are scenarios going to be posted to choose from?
If not, my final tally would be:

Indy - 6
Pitt - 5
NO - 5
Miami - 5

 

[Kenny F'ing Powers]

I'm so mad you jerk don't like Arizona this week.

 

[CantKeepmedown]

guys i'm totally drained so I'll be checking in tomorrow morning and my proposal is we go
 
9/6/4/2 based on the %s this week
 
Kick around anything you like and make your case to others and when I wake up tomorrow and settle in to make the picks around 11, I'm hoping to see some sort of scenario has one out.

If we're doing a 9-6-4-2 spread, I would propose:
 
Miami - 9
NO -6
Baltimore - 4
PItt - 2
 
I'd be open to switching around Balt and Pitt, based on what others think.  I know some were pretty scared of the Pitt pick.  

 

[wade boggs chicken dinner]

I'll bet we could count on one hand the number of people in this thread who can pick the winner ATS 55% or more of the time. We're not going to win because we know football better than the masses. We might, however, be able to solve for the unique problem before us -- maximizing the chance of winning a massive survivor pool with 20ish entries -- better than the other people in that pool.

I'd much rather have people cite Vegas odds as Gospel truth and use them to brainstorm winning pool strategies than have those people go on and on about how they like one 2-1 favorite and not another, based on no-shit factors that are clearly baked into the odds. Those are the posts that don't add value, imo.

I'm going to disagree with this a little bit. I have been arguing that in the early weeks the odds are often off with respect to certain teams and if we can figure this out, we can make some hay. For example, it looks like BUF was underestimated and perhaps SEA was overestimated.

And in fact, if we really believe is this strategy is doable, the LAST thing we should do is to bet along with the general public.

I agree though that our other best bet is to try to figure out a better strategy than others, even though I suspect with a pool this size, a lot of the entries are multiple.