GaMbOL, a fun little experiment

[awayfromjay]

I think you should instead do something like 50% of your BR per day to keep this thread going longer.

[Performify]

If you're really intent on doing this, splitting the money in half each day on two different bets will dramatically reduce your RoR.

i.e. tomorrow place two $2480 bets. Next day place two ~$5000 bets. Day after place two ~10k bets. etc.

You'd make it a lot more likely - in terms of going from an almost impossible shot to a very very very longshot, of course, but still its a massive improvement - to get to your goal.

I've got some game theory text that talks about this someplace but can't dig it up right now.

[dankhank]

if he splits up the bets he will need more winners before he gets to 100k. it's twice as hard.

[ScottieK]

Congrats on your win! This is awesome....I should do this (on a much smaller scale, can't be as baller.)

gogogogogogogo

ScottieK

[Grasshopp3r]

Are you assuming that ruin is certain if he continues with his stated plan? He only has to do roughly 5 more bets to hit $100k if he doubles every time? If he is a 60% favorite, that is 7.8% chance of success (or did I calc it wrong?)

He will exceed the maximum bet size for one site after he hits $10k to $30k, so he will have to split the bets among several sites.

Go DOTTT, you crazy gamboooler. I hope that you do it.

[BobJoeJim]

I think I may follow along on this as well, at a 1% scale (what with my bankroll being one of those that was doubled by the Mansion bet). I missed out on the Oakland game, unfortunately, but I believe you have $4960 right now?

I'll follow all your future picks, starting with $49.60. Here's hoping I can turn it into $1000 for a 50% boost in my post-Pittsburgh bankroll!

[Performify]

[ QUOTE ]
if he splits up the bets he will need more winners before he gets to 100k. it's twice as hard.

[/ QUOTE ]

i guess it depends on how you define "twice as hard". You have to pick more overall winners. But you have significantly less of a risk of going busto, so in that regard (% chance of busto compared to other outcomes) it is "easier" in that regard. Plus it buys a lot more time for him to decide to pull his winnings out of the system and give up the quest [img]/images/graemlins/grin.gif[/img]

OP has to win six bets in a row right now assuming he doesn't use any massive underdogs along the way. The first is down, so he's got five left.

a 50-50 shot at each step would give him a 3% chance of winning five in a row from here. and a 97% chance of Robusto

The odds of hitting six 50-50 in a row would be ~1.5%. A 98.5% chance of Robusto.

assuming he's in fact 50-50, and ignoring juice, starting with 4k and takes the double bet path:

Step one:
WW : 25% to 8k: 2k EV
WL : 50% to 4k: 2k EV
LL : 25% to 0k: 0k EV | 25% Robusto

Step two, WW in round one:
25% to 25% to 16k @ 6.25% : 1k EV |
25% to 50% to 8k: @ 12.5% : 1k EV
25% to 25% to 0k: 0k EV | 6.25% Robusto

Step two, WL in round one:
50% to 25% to 8k @ 12.5% : 1k EV
50% to 50% to 4k @ 25%: 1k EV
50% to 25% to 0k: 0k EV | 12.5% Robusto

LL in round one: 25% Robusto

So we have 4k EV in round 2 just like if we take a 50-50 on our 8k.

But our risk of robusto? 25% with LL in round one, 12.5% with WW>LL, 6.25% with WL>LL == 43.75% of going Robusto by this round. But of course we've only got a 6.25% of winning all four coinflips for WW>WW=16k, a 25% chance of being at 8k, and a 25% chance of being at 4k still.

This is the part I got from the game theory text, again assuming I'm remembering right. any time you can get the same EV outcome with less risk of being broke, its a winner...

Then extrapolating that to the third step:


Step three:

WW | WW | WW: 25% * 25% * 25% to 32k @ 1.5625%: .5k EV
WW | WW | WL : 25% * 25% * 50% to 16k @ 3.125%: .5k EV
WW | WW | LL : 25% * 25% * 25% to 0k @ : 1.5625% chance of Robusto

WW | WL | WW: 25% * 50% * 25% to 8k @ 3.125% : .25k EV
WW | WL | WL : 25% * 50% * 50% to 4k @ 6.25% : .25k EV
WW | WL | LL : 25% * 50% * 25% to 0k @ 3.125% chance of Robusto

WL | WW | WW: 50% * 25% * 25% to 16k @ 3.125% : .5k EV
WL | WW | WL : 50% * 25% * 50% to 8k @ 6.25% : .5k EV
WL | WW | LL : 50% * 25% * 25% to 0k @ 3.125% chance of Robusto

WL | WL | WW : 50% * 50% * 25% to 8k @ 6.25% : .5k EV
WL | WL | WL : 50% * 50% * 50% to 4k @ 12.5% : .5k EV
WL | WL | LL : 50% * 50% * 25% to 0k @ 6.25% chance of Robusto

carry forward:
WW | LL : 6.25% Robusto
WL | LL : 12.5% Robusto
LL first round: 25% Robusto

Chance of Robusto by the third round if you're taking two flips?
25% + 12.5% + 6.25% + 6.25% + 3.125% + 3.125% + 1.5625% = 57.8125%
But giving up that risk costs us EV: we've got a total EV in the third round of 3.5k
A mere 1.5% chance of being at 32k, 6.25% chance of being at 16k, 15.625% chance of being at 8k, 18.75% of being at 4k.

Risk of going Robusto by the third round if we just flip our full roll?
50% first round. 75% by the second round. 87.5% by the third round.

we've got approx 30% more chance to be alive if we split our roll to two bets. And that's just the third round.

I'm not going to run this all the way out to our chances of getting the roll above 100k. But with a ~98% chance of failure on a six-round 50-50 shot, and in a situation where we have a few percentage points of edge at most...

[Performify]

Assume we're 60-40 to win, which i think is probably unrealistic but let's go with it:

no split:

W 60% 8k : 4.8k EV
W 40% Robusto

WW: 60% to 60% to 16k @ 36% : 5.76k EV
WL : 60% to 40% to 0k @ 24% Robusto
carry: 40% Robusto

total 64% busted by second round

WWW : 60% to 60% to 60% to 32k @ 21.6% : 6.912k EV
WWL : 60% to 60% to 40% to 0k @ 14.4% Robusto
WL: 24%
L : 40%

total 78.4% busted by third round


Now splitting in two:

Step one:
WW : 60% to 60% to 8k @ 36% : 2.88k EV
WL : 60% to 40% to 4k @ 48% : 1.92k EV
LL: 40% to 40% to 0k @ 16% Robusto



4.8k EV first round

Step two:
WW | WW : 36% to 36% to 16k @ 12.96% : 2.07k EV
WW | WL : 36% to 48% to 8k: @ 17.28% : 1.38k EV
WW | LL : 36% to 16% to 0k: 0k EV @ 5.76% Robusto

WL | WW : 48% to 25% to 8k @ 12.5% : 1k EV
WL | WL : 48% to 48% to 4k @ 23.04%: .9216k EV
WL | LL 48% to 16% to 0k: 0k EV @ 7.68% Robusto

carry LL in round one: 16% Robusto

So we have 5.3716k EV in round 2

But our risk of robusto? 29.44% But of course we've only got a 12.96% of winning all four for WW>WW=16k, a 29.78% chance of being at 8k, and a 23.04% chance of being at 4k still.

Then extrapolating that to the third step:

Step three:

WW | WW | WW: 36% * 36% * 36% to 32k @ 4.6656%:
WW | WW | WL : 36% * 36% * 48% to 16k @ 6.2208%:
WW | WW | LL : 36% * 36% * 16% to 0k @ : 2.0736% chance of Robusto

WW | WL | WW: 36% * 48% * 36% to 8k @ 6.2208% :
WW | WL | WL : 36% * 48% * 48% to 4k @ 8.2944% :
WW | WL | LL : 36% * 48% * 16% to 0k @ 2.7648% chance of Robusto

WL | WW | WW: 48% * 36% * 36% to 16k @ 6.2208% :
WL | WW | WL : 48% * 36% * 48% to 8k @ 8.2944% :
WL | WW | LL : 48% * 36% * 16% to 0k @ 2.7648% chance of Robusto

WL | WL | WW : 48% * 48% * 36% to 8k @ 8.2944% :
WL | WL | WL : 48% * 48% * 48% to 4k @ 11.0958% :
WL | WL | LL : 48% * 48% * 16% to 0k @ 3.6864% chance of Robusto

carry forward:
WW | LL : 5.76% Robusto
WL | LL : 7.68% Robusto
LL first round: 16% Robusto

Total chance of busting by the third round: less than 41%.
32k @ ~5%:
16k @ ~12.5%
8K @ ~21%
4k @ ~19%

Chance of busting third round straight @ 60% edge is close to 80%. we've got a 20% chance of having 32k though.



hmmm. This may have been entirely pointless [img]/images/graemlins/grin.gif[/img]

[dankhank]

[ QUOTE ]


hmmm. This may have been entirely pointless [img]/images/graemlins/grin.gif[/img]

[/ QUOTE ]

well, my comment was meant to be sarcastic

[RivaLiva]

For Thursdays game:

I would probably choose
NY Yankees -1.5 at home vs Tampa Bay devil rays
Wang has pitched well the 2nd half of the season and the Yanks are hot.

Or

West Virginia moneyline vs Maryland
This won't double the bankroll because it's probably -200 at this point or higher I haven't looked.