Top Grossers: 8/31/18

[Kyle]

But Mike wins the season.

[poorpete]

[Kyle]

Actuals for Crazy Rich Asians- $21,964,345

Results:
poorpete- 0
Mike- 3
Kyle- 4
Zombie- 5
buckett- 6
Zen- 7

Final results for season:
Mike-156
buckett- 158
Zombie- 165
Zen- 179
Kyle- 192
poorpete- 220

[Mike]

Yay me! My moment in the sun.

[poorpete]

[Zen]

Congrats Mike!

And, so the info is all in one thread:

Summer 2018 Weekly Win Results: (Final)
Zen: 5
buckett: 3
Mike: 3
poorpete: 2
Kyle: 1
Zombie: 1
mimekiller: 0
Dan: 0

Which proves beyond a shadow of a doubt that weekly wins matter naught... or at least minimally...

[Mike]

It's not that weekly wins don't matter, it's that they are a different game. There's a completely different risk assessment to be made if you are playing win/lose versus playing for margin of error. I like to think that I play the same either way, but the last couple weeks have shown me I don't always.

[Zen]

Well, if you go for the knockout, you can sometimes get yourself out of position and get caught with a hay-maker, which puts you behind in the "golf scoring", even if you've won more rounds than your opponent...

[Mike]

Exactly. My point being that you and I are playing similar but different games. They just happen to be played simultaneously on the same field.

[Mike]

I've always wondered about crowdsourcing this game. That is, what would happen if you waited until the last minute each week, and then played whatever number was closest to the average of all the guesses so far.

Turns out, it would suck. I added it up for this season, and it would have scored 209. Sad.

[poorpete]

Damn Deadpool 2

[Zen]

[Mike]

How much WE suck is irrelevant. The thing that interested me is that playing the average every week resulted in a final score that was WAY below average. The average score was 178, ranging from the best at 156 to the worst at 220. But playing the average each week, lands you at 209, very nearly the worst.

This is even more significant when you realize that I'm only playing the average of the players who actually made a guess. People who missed a week took a penalty that would have dragged the final scores down while not affecting my "average" player.

On the other hand, about half the time, my fictional player couldn't take the average, because that guess was already taken, and on a tight week, it might have had to shift by 2 or 3 million to find an open number.

[Zen]

So this suggests that extremes may be the better strategy?

[Mike]

I don't know. I'd have to crunch the numbers, but I feel like Pete was the biggest outlier most weeks, and he finished dead last. You were probably second, and you finished dead average. Don't know what that tells us.

[Zen]

Well, I win a lot of weeks, but when I lose, I lose spectacularly... so I'd say it balances out... (Also, I seem to have a tendency to win on weeks when everyone takes it on the chin... so it doesn't seem like it is much of a win at the time...) So, yeah... extremes probably lead to that kind of thing...

[Mike]

Nevermind. I'm an idiot... as usual.

I plugged stuff into a spreadsheet, and it turns out that if I played the average every week, my final score would actually have been 171. Which would have put it in 4th place, right in the middle of the pack.

If I could actually play the average and not be constricted to whole numbers that haven't been taken yet, it would have been 166. If I could play the median instead, it would have jumped to 162 and 3rd place.

Lesson learned, playing averages won't win you any prizes, but on a given week when you are pressed for time and/or haven't got a good feel for the movies that week, playing the average will at least not put you much behind as a holdover. A lesson which we probably could have guessed from common sense without an hour of playing with spreadsheets.

[Mike]

More number geeking: who takes the most risks?

I looked at how far off of the average each player was each week (on average)...

poorpete -- 5.07
Zen -- 3.73
Kyle -- 3.51
Mike -- 2.62
dan -- 2.39
buckett -- 1.73
Zombie -- 1.63

I was right that poorpete was the biggest outlier, with Zen a distant second and Kyle right behind. I think what we learned here is that Zombie and buckett are the steadiest players and will likely finish near the top most seasons. I take moderate risks, which in this case happened to pay off for me, landing me in the top spot, but previously has gone south on me and put me 2nd to last. I think for me it was more luck than skill that I happened to come out on top this time.

[Zombie]

Kyle's number is actually pretty impressive considering he starts it off every week.

[Mike]

I agree, the fact that Kyle ALWAYS makes a cold guess makes him hard to pin down. The rest of us have the advantage of playing strategically when we feel like it, or allowing the other guesses to tell us when we're completely off base. Not for Kyle.