## Poll

 Go for a 53%-55% theoretical win-rate No votes (0%) ... 56%-58% ... No votes (0%) ... 59%-61% ... No votes (0%) I think the betting limits would have to be small No votes (0%) "Cheaters never prosper" 1 vote (33.33%) "Stick to the lotto bets" 2 votes (66.66%) Other (please explain) No votes (0%)

3 members have voted

ksdjdj
Joined: Oct 20, 2013
• Posts: 909
May 28th, 2020 at 1:22:22 AM permalink
I only just noticed this, but about two months ago an Australian Sportsbook offered "...odds on replays of past AFL and NRL matches...". (see link below for more details)
https://news.sportsbet.com.au/afl/sportsbet-opens-betting-on-classic-footy-matches/

Anyway, below are the questions for this thread:

1. Would you try to "mathematically beat" this, if you could?
2. If you answered yes to question 1., how would you do it, and why?
3. If you answered no to question 1., then explain the reason(s) why ? (eg "betting limits would probably have been too small", "i consider this cheating." and/or "other reasons" ...)

----
hypothetical (what I would consider doing, from a "+ EV point of view, whilst also trying to limit my chances of getting caught" ):

Assumptions:
. Betting "against the spread" /ats.
. \$1.90 are the typical odds on offer with this book, when betting ats.
. The 'book will ban you and confiscate your winnings from any of the "classic matches" if your win/strike-rate is above 2/3 (66.66..%***) after 10 (or more) games.

*** (Update about 215 am): The above assumption is just a guess (if you disagree that it is a "plausible assumption", please provide an alternative).

Steps I would take, to have "a good chance of +EV", and also "to reduce the risk of getting banned":
1. Select a "classic game" you are about to have a bet on.
2. Open a spreadsheet that has a "random number generator" / rng, with values between 0.01-10
3 (a)Bet on the "winning team ats", any time the rng showed a number between 9.01 and 10.
3 (b)Bet on a "random team ats", any time the rng showed a number between 0.01 and 9.00 (made random by flipping a coin, eg "heads = a bet on the home team...")
4. Repeat steps 1 to 3, if/when there are more "classic games" to bet on.

Using the above steps my "long-term" EV should be +4.5% ### (see "proof" below)

"Proof":
.You will " bet on the guaranteed winner ats" 10% of the time (long-term), because the rng should show a number between 9.01 and 10, "10% of the time" on average.
. 90% of the time you will bet on the "coin flip bets"
.you are expected to win 45% of the "coin flip bets",
.you are expected to lose 45% of the "coin flip bets"

"Expected Win rate" = "...guaranteed winner" + "winning coin flip bets"
= 10% + 45% = 55%
"Expected win rate" x "odds" = RTP
= 55% x 1.90 = 104.5%
EV = RTP - 100% = 4.5%

Therefore the EV = 4.5%. ###

### (Important): This EV figure doesn't take into account that your winnings will be confiscated, if your strike-rate is above 2/3 after 10 or more games"

---
Hope you find this thread interesting.

Spelling not thoroughly checked, but math should be correct.
Last edited by: ksdjdj on May 28, 2020
SOOPOO
Joined: Aug 8, 2010
• Posts: 7804
May 28th, 2020 at 6:09:20 AM permalink
This is silly. I'd be 6/10 after 10 games. 13/20 after 20 games. 20/30 after 30 games. Gotta lose game 31.... but 26/40 after 40 games.

Edit... if I had to check a box or something saying I did not know the results of the match I would not participate.
DRich
Joined: Jul 6, 2012
• Posts: 7290
May 28th, 2020 at 7:14:54 AM permalink
I think they would be smart to do a parimutual style betting on these. Take all bets, skim off 5% for the sportsbook, payback the other 95% to winners.
Living longer does not always infer +EV
ThatDonGuy
Joined: Jun 22, 2011
• Posts: 4683
May 28th, 2020 at 7:55:59 AM permalink
Quote: DRich

I think they would be smart to do a parimutual style betting on these. Take all bets, skim off 5% for the sportsbook, payback the other 95% to winners.

Does the track's cut stop someone that knows the result of a horse race in advance from betting on it?

I like these quotes:
“Obviously, we will be asking customers to only wager on matches they do not already know the result of.”
"Sportsbet requests an honour system from customers who are familiar with the outcome of the market."

Maybe if they chose a point in the match in advance (unannounced, of course) when the result would be determined, it would work, but if it is always based on the full 80 minutes (rugby) / 80-100 minutes (AFL - at some point, quarters were changed from 25 minutes to 20, although with 25, the clock kept running after scores and when the ball went out of bounds; recently they were reduced again, to 16), this is asking for the book to be taken for quite a bit of money. This won't last very long under that format, if it even starts.
billryan
Joined: Nov 2, 2009
• Posts: 10436
May 28th, 2020 at 8:43:11 AM permalink
Charities in the US had "Nights at the races" where people bet on old horse races that had been run years before.
I look at this as if ESPN is going to broadcast a Yankee- Red Sox game tonite and you can bet on it. You don't know which game it is, just a game from their long rivalry. You might know the final score of every game in the series but that doesn't help unless you know exactly which game they will show.
ThatDonGuy
Joined: Jun 22, 2011
• Posts: 4683
May 28th, 2020 at 10:34:29 AM permalink
Quote: billryan

Charities in the US had "Nights at the races" where people bet on old horse races that had been run years before.
I look at this as if ESPN is going to broadcast a Yankee- Red Sox game tonite and you can bet on it. You don't know which game it is, just a game from their long rivalry. You might know the final score of every game in the series but that doesn't help unless you know exactly which game they will show.

The video says that bets will be taken only on the matches aired on TV, and those will have their dates and teams specified in advance, so it shouldn't be that hard to determine the winner in advance.

I wonder how many people will have accomplices, or second accounts, where they bet a smaller amount on the loser so it doesn't look suspicious. Then again, that could backfire if the book notices, say, a lot of \$200 bets on the winner and an almost equal number of \$100 bets on the loser.
billryan
Joined: Nov 2, 2009
• Posts: 10436
May 28th, 2020 at 10:52:31 AM permalink
Quote: ThatDonGuy

The video says that bets will be taken only on the matches aired on TV, and those will have their dates and teams specified in advance, so it shouldn't be that hard to determine the winner in advance.

I wonder how many people will have accomplices, or second accounts, where they bet a smaller amount on the loser so it doesn't look suspicious. Then again, that could backfire if the book notices, say, a lot of \$200 bets on the winner and an almost equal number of \$100 bets on the loser.

That doesn't really make sense, but hopefully, someone beats them silly.
ThatDonGuy
Joined: Jun 22, 2011
• Posts: 4683
May 28th, 2020 at 10:55:35 AM permalink
Quote: billryan

That doesn't really make sense, but hopefully, someone beats them silly.

If everybody bets on the winner, they stop after one match. If it appears balanced, they keep it going, and the bettors make their money a bit at a time - at least until somebody gets tired of it and makes the one big bet that causes the book to come to its senses.

For all I know, there's nothing to stop the book from cancelling all bets before the match begins - which, if they can, they would certainly do if it stood to lose a lot of money.
TomG
Joined: Sep 26, 2010
• Posts: 2244
May 28th, 2020 at 1:05:16 PM permalink
Quote: SOOPOO

This is silly. I'd be 6/10 after 10 games. 13/20 after 20 games. 20/30 after 30 games. Gotta lose game 31.... but 26/40 after 40 games.

Edit... if I had to check a box or something saying I did not know the results of the match I would not participate.

Also, bet the minimum on the losses and bet the max on the wins. Or if you don't expect the offers to last long, just bet the winning side for the max nine times and cash out.

If they have the authority and willingness to steal someones money, and we're just guessing on what might cause them to do that, then it's just gambling no matter how it's played.
TomG
Joined: Sep 26, 2010