Parrondo's Paradox in Sports Betting (multiple -ev wagers creating +ev) Yah sounds ridiculous, But its interesting promise!

[aayjay]

Is it possible multiple -EV wagers can combine into some kind of +EV wager(s)?

sounds retarded i know...

Could there be some combination of -ev wagers that create +ev wagers? Obviously correlated parlays would do that, but does not seem like that's what there going for here.

Parrondo's paradox, a paradox in game theory, has been described as: A combination of losing strategies becomes a winning strategy. It is named after its creator, Juan Parrondo, who discovered the paradox in 1996.

more info on Parrondo's paradox:

https://en.wikipedia.org/wiki/Parron...ox%20in%201996.

Found this on twitter, the "system" has been talked about for more than a few years so there is some back data on it. And the original thread is still going strong. Feel free to read through it, see what you think, think this is possible?

Anyhow I also copied / pasted so you don't have to visit twitters shit hole to read the thread. And also below the thread is the link to the original NHL thread, there is also a mlb thread that came first but I was to lazy to get it right now:

The twitter thread I found was written by:

Andrew Mack
@Gingfacekillah

Here is a link to the twitter thread:

Embedded Tweet

https://twitter.com/Gingfacekillah/status/1635711305641529346

And here it is copy/pasted:

Parrondo's Paradox in Sports Betting ��

Parrondo's Paradox refers to the concept that 2 losing games can sometimes be combined to producing a winning result. This strategy might not be a perfect example, but I think you'll see what I mean.

Backstory: back in 2014, Danrules24 posted on the covers SB forum about a parlay martingale strategy applied to both MLB and NHL. On its surface it appeared to be completely degenerate - but the results over a few seasons were intriguing.

The strategy takes the two strongest road teams and parlays them together, then sequences bets in a martingale structure for 8 bets. The betting sequence itself comes from a book called "12 Ways to Beat Your Bookie" by Tony Stoffo (basically a book of martingale strats).

Now on its surface, this sounds like degeneracy 101. Pure gamblooooor stuff. Each strategy should be a long term net loser, but Danrules results were profitable over several seasons.

Maybe it's worth a backtest.

I grabbed some data for several NHL seasons. One tweak I added was to select the two best-priced road favourites (highest payoff while still being shorter than 2.00 decimal odds). If there were more than 2 with the same price, I randomly selected the 2.

I also changed the bet sequence amounts to pure martingale - i.e. doubling the amount every time rather than the multipliers recommended by Stoffo.

After that, it was time to run a backtest.

Result: In 3 seasons on NHL, the strategy appeared profitable. The return distribution is negatively skewed, showing small methodical series wins followed by blowout losses. Interestingly, even shuffling the bets up, it still appears to be a winning strategy.

Is it likely persist? Maybe, but I personally wouldn't bet this because the uncertainty around the blowouts means it's not for the feint of heart. It's also unclear to me what the edge is, other than perhaps betting that NHL games are more random than the market thinks.

I'm not even sure that this is really an example of Parrondo's paradox tbh, but it reminded me of the concept when I was thinking it over.

Either way, I hope you found this interesting. Thanks to
@neilgcurrie
for assistance with the backtest on this one.

/fin

MLB discussion (not original):
https://www.covers.com/forum/systems...stem-102418263

NHL discussion:
https://www.covers.com/forum/systems...stem-101980215


Twitter thread:

Embedded Tweet

https://twitter.com/Gingfacekillah/status/1635711305641529346

[sonatine]

christ i bet this would crush with options...

[BCR]

Is it possible multiple -EV wagers can combine into some kind of +EV wager(s)?

sounds retarded i know...

Could there be some combination of -ev wagers that create +ev wagers? Obviously correlated parlays would do that, but does not seem like that's what there going for here.

Parrondo's paradox, a paradox in game theory, has been described as: A combination of losing strategies becomes a winning strategy. It is named after its creator, Juan Parrondo, who discovered the paradox in 1996.

more info on Parrondo's paradox:

https://en.wikipedia.org/wiki/Parron...ox%20in%201996.

Found this on twitter, the "system" has been talked about for more than a few years so there is some back data on it. And the original thread is still going strong. Feel free to read through it, see what you think, think this is possible?

Anyhow I also copied / pasted so you don't have to visit twitters shit hole to read the thread. And also below the thread is the link to the original NHL thread, there is also a mlb thread that came first but I was to lazy to get it right now:

The twitter thread I found was written by:

Andrew Mack
@Gingfacekillah

Here is a link to the twitter thread:

Embedded Tweet

https://twitter.com/Gingfacekillah/status/1635711305641529346

And here it is copy/pasted:

Parrondo's Paradox in Sports Betting ��

Parrondo's Paradox refers to the concept that 2 losing games can sometimes be combined to producing a winning result. This strategy might not be a perfect example, but I think you'll see what I mean.

Backstory: back in 2014, Danrules24 posted on the covers SB forum about a parlay martingale strategy applied to both MLB and NHL. On its surface it appeared to be completely degenerate - but the results over a few seasons were intriguing.

The strategy takes the two strongest road teams and parlays them together, then sequences bets in a martingale structure for 8 bets. The betting sequence itself comes from a book called "12 Ways to Beat Your Bookie" by Tony Stoffo (basically a book of martingale strats).

Now on its surface, this sounds like degeneracy 101. Pure gamblooooor stuff. Each strategy should be a long term net loser, but Danrules results were profitable over several seasons.

Maybe it's worth a backtest.

I grabbed some data for several NHL seasons. One tweak I added was to select the two best-priced road favourites (highest payoff while still being shorter than 2.00 decimal odds). If there were more than 2 with the same price, I randomly selected the 2.

I also changed the bet sequence amounts to pure martingale - i.e. doubling the amount every time rather than the multipliers recommended by Stoffo.

After that, it was time to run a backtest.

Result: In 3 seasons on NHL, the strategy appeared profitable. The return distribution is negatively skewed, showing small methodical series wins followed by blowout losses. Interestingly, even shuffling the bets up, it still appears to be a winning strategy.

Is it likely persist? Maybe, but I personally wouldn't bet this because the uncertainty around the blowouts means it's not for the feint of heart. It's also unclear to me what the edge is, other than perhaps betting that NHL games are more random than the market thinks.

I'm not even sure that this is really an example of Parrondo's paradox tbh, but it reminded me of the concept when I was thinking it over.

Either way, I hope you found this interesting. Thanks to
@neilgcurrie
for assistance with the backtest on this one.

/fin

MLB discussion (not original):
https://www.covers.com/forum/systems...stem-102418263

NHL discussion:
https://www.covers.com/forum/systems...stem-101980215


Twitter thread:

Embedded Tweet

https://twitter.com/Gingfacekillah/status/1635711305641529346

I read the Danrulez NHL thread. Or at least 3 seasons of it. Interesting. He never loved the -210/-170 stuff. Occasionally he’d throw a PL in there. I just looked at NHL tonight and theoretically I’d be betting Devils/Bruins parlay for even money as they’re -250 and -230. Didn’t see many of those. Is that an anomaly? My problem is I wouldn’t know where to throw a PL in not following sport.

[aayjay]

Is it possible multiple -EV wagers can combine into some kind of +EV wager(s)?

sounds retarded i know...

Could there be some combination of -ev wagers that create +ev wagers? Obviously correlated parlays would do that, but does not seem like that's what there going for here.

Parrondo's paradox, a paradox in game theory, has been described as: A combination of losing strategies becomes a winning strategy. It is named after its creator, Juan Parrondo, who discovered the paradox in 1996.

more info on Parrondo's paradox:

https://en.wikipedia.org/wiki/Parron...ox%20in%201996.

Found this on twitter, the "system" has been talked about for more than a few years so there is some back data on it. And the original thread is still going strong. Feel free to read through it, see what you think, think this is possible?

Anyhow I also copied / pasted so you don't have to visit twitters shit hole to read the thread. And also below the thread is the link to the original NHL thread, there is also a mlb thread that came first but I was to lazy to get it right now:

The twitter thread I found was written by:

Andrew Mack
@Gingfacekillah

Here is a link to the twitter thread:

Embedded Tweet

https://twitter.com/Gingfacekillah/status/1635711305641529346

And here it is copy/pasted:

Parrondo's Paradox in Sports Betting ��

Parrondo's Paradox refers to the concept that 2 losing games can sometimes be combined to producing a winning result. This strategy might not be a perfect example, but I think you'll see what I mean.

Backstory: back in 2014, Danrules24 posted on the covers SB forum about a parlay martingale strategy applied to both MLB and NHL. On its surface it appeared to be completely degenerate - but the results over a few seasons were intriguing.

The strategy takes the two strongest road teams and parlays them together, then sequences bets in a martingale structure for 8 bets. The betting sequence itself comes from a book called "12 Ways to Beat Your Bookie" by Tony Stoffo (basically a book of martingale strats).

Now on its surface, this sounds like degeneracy 101. Pure gamblooooor stuff. Each strategy should be a long term net loser, but Danrules results were profitable over several seasons.

Maybe it's worth a backtest.

I grabbed some data for several NHL seasons. One tweak I added was to select the two best-priced road favourites (highest payoff while still being shorter than 2.00 decimal odds). If there were more than 2 with the same price, I randomly selected the 2.

I also changed the bet sequence amounts to pure martingale - i.e. doubling the amount every time rather than the multipliers recommended by Stoffo.

After that, it was time to run a backtest.

Result: In 3 seasons on NHL, the strategy appeared profitable. The return distribution is negatively skewed, showing small methodical series wins followed by blowout losses. Interestingly, even shuffling the bets up, it still appears to be a winning strategy.

Is it likely persist? Maybe, but I personally wouldn't bet this because the uncertainty around the blowouts means it's not for the feint of heart. It's also unclear to me what the edge is, other than perhaps betting that NHL games are more random than the market thinks.

I'm not even sure that this is really an example of Parrondo's paradox tbh, but it reminded me of the concept when I was thinking it over.

Either way, I hope you found this interesting. Thanks to
@neilgcurrie
for assistance with the backtest on this one.

/fin

MLB discussion (not original):
https://www.covers.com/forum/systems...stem-102418263

NHL discussion:
https://www.covers.com/forum/systems...stem-101980215


Twitter thread:

Embedded Tweet

https://twitter.com/Gingfacekillah/status/1635711305641529346

I read the Danrulez NHL thread. Or at least 3 seasons of it. Interesting. He never loved the -210/-170 stuff. Occasionally he’d throw a PL in there. I just looked at NHL tonight and theoretically I’d be betting Devils/Bruins parlay for even money as they’re -250 and -230. Didn’t see many of those. Is that an anomaly? My problem is I wouldn’t know where to throw a PL in not following sport.

I messed around w/ it a little bit during baseball, my own version was just to parlay the 2 or 3 road favs closest to even $, so like lets just say NYY -112 w/. NYM -135 , something like that, dont think you see alot of road favs higher than like -150ish in baseball. it was profitable for the short time i played with it & i never martingaled but gonna maybe add some version, the difference is pretty significant

[BCR]

I read the Danrulez NHL thread. Or at least 3 seasons of it. Interesting. He never loved the -210/-170 stuff. Occasionally he’d throw a PL in there. I just looked at NHL tonight and theoretically I’d be betting Devils/Bruins parlay for even money as they’re -250 and -230. Didn’t see many of those. Is that an anomaly? My problem is I wouldn’t know where to throw a PL in not following sport.

I messed around w/ it a little bit during baseball, my own version was just to parlay the 2 or 3 road favs closest to even $, so like lets just say NYY -112 w/. NYM -135 , something like that, dont think you see alot of road favs higher than like -150ish in baseball. it was profitable for the short time i played with it & i never martingaled but gonna maybe add some version, the difference is pretty significant

Update if you try it. I found it interesting, but obviously no one feels like line shopping all year to make $1000. But you also can’t start very nigh because it gets to game 6 and 7 too often. You’d need to be running 20 different series simultaneously in multiple sports, but there aren’t enough high volume sports. Interesting though that it seems to work, just isn’t scalable from what I can tell.

-ignore the emojis. I’m always on my phone and fat finger them by accident and can’t edit them out. I’m certainly not giving your post a thumbs down. I find it interesting.

[BCR]

Most of the NBA back tests failed. I tried a small variation on what they tried. Including Orlando twice is definitely not a true test of the system, actually did it by accident. Obviously spread complicates system.
I aimed for that 2-1 to 3-1 payout. I only included amounts bet because this is just an experiment and $30 range seems to be about as high as you can go and roll it over for a series of 6-7x. Anything higher would get you above bet limit.

If it shows any unit success, I’ll roll it over to wagering thread, not trying to create a separate thread, but I doubt it will with NBA load management and I’ll abandon it after a week and don’t want to clutter serious hockey focused thread. I just don’t follow NHL, so don’t have the insight to lay off certain teams like the Dan guy did in NHL.

Parlay (2 Picks)0 of 2 settled
Oct 25 • 2023 At 01:08:44 AM
Ticket ID: 6344424
Point Spread: OKC Thunder +1.0 @ -113
OKC Thunder @ CHI Bulls
Moneyline: ORL Magic @ -167
HOU Rockets @ ORL Magic
Odds:
+202
Wager:
$33.00
Pot. Payout:
$99.79

Parlay (2 Picks)0 of 2 settled
Oct 25 • 2023 At 12:54:06 AM
Ticket ID: 6344393759
Point Spread: ORL Magic -4.0 @ -110
HOU Rockets @ ORL Magic
Moneyline: MIL Bucks @ -225
PHI 76ers @ MIL Bucks
Odds:
+177
Wager:
$30.00
Pot. Payout:
$83.09

[aayjay]

[QUOTE=BCR;1101581]

-ignore the emojis. I’m always on my phone and fat finger them by accident and can’t edit them out. I’m certainly not giving your post a thumbs down. I find it interesting.

funny, the emojis did confuse, me although i thought it was more of a emotional response like = "WOW really?" and then thumbs down = "damn not worth it" , so even though i was confused didn't take it poorly.

Something i found interesting is that some of the guys going over this are smart cookies, real sharp data guys, and so far don't think i've seen anyone really show it isn't working, even with randomization.

At least for baseball think its sort of like placing a season long wager that theyre will be more variance in home team loses than the market prices in. I actually think i saw someone post that sort of a message somewhere so I am probably just-rehashing someone elses perspective here. Not sure, been a few months since i read through these posts, was going to post here earlier but just wasn't enough sharp sports discussion, hopefully baseball/fotball/basketball/hockey/ncaa can get some sharp sports talk goin'

[aayjay]

Most of the NBA back tests failed. I tried a small variation on what they tried. Including Orlando twice is definitely not a true test of the system, actually did it by accident. Obviously spread complicates system.
I aimed for that 2-1 to 3-1 payout. I only included amounts bet because this is just an experiment and $30 range seems to be about as high as you can go and roll it over for a series of 6-7x. Anything higher would get you above bet limit.

If it shows any unit success, I’ll roll it over to wagering thread, not trying to create a separate thread, but I doubt it will with NBA load management and I’ll abandon it after a week and don’t want to clutter serious hockey focused thread. I just don’t follow NHL, so don’t have the insight to lay off certain teams like the Dan guy did in NHL.

Parlay (2 Picks)0 of 2 settled
Oct 25 • 2023 At 01:08:44 AM
Ticket ID: 6344424
Point Spread: OKC Thunder +1.0 @ -113
OKC Thunder @ CHI Bulls
Moneyline: ORL Magic @ -167
HOU Rockets @ ORL Magic
Odds:
+202
Wager:
$33.00
Pot. Payout:
$99.79

Parlay (2 Picks)0 of 2 settled
Oct 25 • 2023 At 12:54:06 AM
Ticket ID: 6344393759
Point Spread: ORL Magic -4.0 @ -110
HOU Rockets @ ORL Magic
Moneyline: MIL Bucks @ -225
PHI 76ers @ MIL Bucks
Odds:
+177
Wager:
$30.00
Pot. Payout:
$83.09

how did you choose the teams to parlay ? i think you wanna try to stick to -200 or better, dont think super heavy favs work as well, but thats just from reading the threads not from testing myself.

also why going w/ spreads? maybe you read more than i did or have some basketball insight but ml seems the play for this imo

[aayjay]

can someone get Druff in here, hes a smart Jewish data guy, wonder if he thinks its bad, terrible, or worse lol

[BCR]

I’m certainly not a sharp. I’m a very good rec poker player, but not a sports sharp. Those here that are probably talk amongst themselves a little more than publicly. Hence why I’m not cluttering up that thread that’s primarily focused on hockey.

I should have started it last night. I had both winners and was pretty convinced Denver would win straight up, and was just a lean that Phx would win. The ML went up when Beal got announced out, but given I feel Beal is a losing player with his defense, it made me like the Phoenix side more. Had I wrapped a Denver ML with below, it likely would have been like +280/300 ish. Can’t recall exactly what Denver ML was.


Single
@
+125
Won
Oct 24 • 2023 At 09:57:02 PM
Ticket ID: 6344024971
Moneyline: PHO Suns
PHO Suns @ GS Warriors


I’m thinking the NBA version would be anchor two parlays around 2 speculative + money picks and a game you’re rather confident is at least a ML winner. You’re going to be wrong and lose that anchor game occasionally, but win it most of the time as it’s going to be a -200 type favorite.

Either way, I hit those first 2 games for real bets and figured I’d plow the profits into this experiment. Much cheaper stakes than I normally bet, and I’ve never martingaled any type of sports wager, at least intentionally. Only when chasing losses back in the bookie on credit era.

[BCR]

Most of the NBA back tests failed. I tried a small variation on what they tried. Including Orlando twice is definitely not a true test of the system, actually did it by accident. Obviously spread complicates system.
I aimed for that 2-1 to 3-1 payout. I only included amounts bet because this is just an experiment and $30 range seems to be about as high as you can go and roll it over for a series of 6-7x. Anything higher would get you above bet limit.

If it shows any unit success, I’ll roll it over to wagering thread, not trying to create a separate thread, but I doubt it will with NBA load management and I’ll abandon it after a week and don’t want to clutter serious hockey focused thread. I just don’t follow NHL, so don’t have the insight to lay off certain teams like the Dan guy did in NHL.

Parlay (2 Picks)0 of 2 settled
Oct 25 • 2023 At 01:08:44 AM
Ticket ID: 6344424
Point Spread: OKC Thunder +1.0 @ -113
OKC Thunder @ CHI Bulls
Moneyline: ORL Magic @ -167
HOU Rockets @ ORL Magic
Odds:
+202
Wager:
$33.00
Pot. Payout:
$99.79

Parlay (2 Picks)0 of 2 settled
Oct 25 • 2023 At 12:54:06 AM
Ticket ID: 6344393759
Point Spread: ORL Magic -4.0 @ -110
HOU Rockets @ ORL Magic
Moneyline: MIL Bucks @ -225
PHI 76ers @ MIL Bucks
Odds:
+177
Wager:
$30.00
Pot. Payout:
$83.09

how did you choose the teams to parlay ? i think you wanna try to stick to -200 or better, dont think super heavy favs work as well, but thats just from reading the threads not from testing myself.

also why going w/ spreads? maybe you read more than i did or have some basketball insight but ml seems the play for this imo

Yeah, I didn’t think it though that much for $60. Just played around a little. I just took a bet I liked with spread, Orlando -4, and used it with ML and spread. And then took OKC, who I liked here anyway. Honestly, if I wanted to cancel bet and realign, I might lay off the Bucks. They are a bit overpriced on the ML bet. I wasn’t thinking about the growing pains when you pair 2 stars. I just felt like trying the Martingale aspect of it, but get me out to 6 or 7 steps, and I doubt I’m firing 4 figures on an NBA parlay. I’ll probably experiment up to $750 or something in that range. That’s about my tolerance for an NBA parlay.

Being opening night, I just am riding with bets on o/u type numbers I projected for these teams and relative strength. I’m more betting against Houston and Chicago than for OKC and Orlando, albeit both those teams are headed in the right direction.

[BCR]

Well that went according to script as well as possible. I thought about it a little after and made a few adjustments. Didn’t matter much as everything I liked won by 30 and 20. I thought Bucks game was last night originally. Can’t martingale shit with a two day parlay.



I actually cashed out of both as I had bet $33 on the one, and that kind of defeats the purpose of the $30 unit. Plus OKC was up to +2.5. Cost me a few $. Nice part of wagers this small.

Same teams for the most part, just added Toronto in place of Milwaukee since they weren’t playing and I liked Toronto also. Also hit a fairly large wager unrelated to this experiment on Orlando as I was very confident in that spot. Took it up to 5.5 for +102 on my personal bet as juice was -114. Should have taken it to 20 and it would have been comfortable.

Parlay (2 Picks)Won
Oct 25 • 2023 At 03:19:39 PM
Ticket ID: 6347557164
Moneyline: TOR Raptors @ -109
MIN Timberwolves @ TOR Raptors
Point Spread: ORL Magic -4.5 @ -113
HOU Rockets @ ORL Magic
Odds:
+263
Wager:
$30.00
Payout:
$108.86

Parlay (2 Picks)Won
Oct 25 • 2023 At 03:17:43 PM
Ticket ID: 6347564506
Moneyline: ORL Magic @ -186
HOU Rockets @ ORL Magic
Point Spread: OKC Thunder +2.5 @ -109
OKC Thunder @ CHI Bulls
Odds:
+196
Wager:
$30.00
Payout:
$88.70




Single @
+102
Won
Oct 25 • 2023 At 03:24:04 PM
Ticket ID: 6347604783
Point Spread: ORL Magic -5.5
HOU Rockets @ ORL Magic

Cash outs I made before games started. Should have just kept them as they’d be winners also.

$28.43 Parlay (2 Picks)Cash Out confirmed
Oct 25 • 2023 At 01:08:44 AM
Ticket ID: 6344424120
Point Spread: OKC Thunder +1.0 @ -113
OKC Thunder @ CHI Bulls
Moneyline: ORL Magic @ -167
HOU Rockets @ ORL Magic
Odds:
+202
Wager:
$33.00
Payout:
$28.43

Parlay (2 Picks)Cash Out confirmed
Oct 25 • 2023 At 12:54:06 AM
Ticket ID: 6344393759
Point Spread: ORL Magic -4.0 @ -110
HOU Rockets @ ORL Magic
Moneyline: MIL Bucks @ -225
PHI 76ers @ MIL Bucks
Odds:
+177
Wager:
$30.00
Payout:
$29.01




So $137.56 profit on first night minus .99 for first buy out, and 4.57 for second. $132.00 profit on first night. +4.4 units

Nothing tonight with 2 games. We will play with this again Friday. Need to think about this experiment a little more as NBA requires more handicapping than just blindly betting 2 best road teams. That gets you killed in NBA as a team like Golden State who was 44-38 last year was like 34-7 at home. Home court more valuable in NBA for decent teams.

[BCR]

Man, I’ve looked through these slates. I’ll bet 2 $30 parlays on Toronto-OKC and Knicks -GS

Those are +389 and +346. That’s the kind of return you need for this to work. I don’t dislike any of those teams shot at winning, but could see long stretches where you go 1 for 2 and lose. I think homecourt and crowds influencing refs might have more influence in NBA, plus load mgmt.

Those are more similar to what they did in those threads.

Personally for myself, I’m betting a SA-Den ML parlay, but +186 isn’t enough value for this experiment. Once you got 3 bets deep, even with martingale, you couldn’t get enough value.

Despite sweeping all my bets first round, what I did were more standard parlays. These 2 are more similar to what they did over on covers.

I’ll probably wait until MLB starts again. It is interesting though, and I liked reading the experiment. There is something there. I just doubt NBA is plausible.

Lines shifted a little by time I bet.


Parlay (2 Picks)0 of 2 settled
Oct 27 • 2023 At 12:14:11 PM
Ticket ID: 6356580747
Moneyline: NY Knicks @ -108
NY Knicks @ ATL Hawks
Moneyline: OKC Thunder @ +135
OKC Thunder @ CLE Cavaliers
Odds:
+354
Wager:
$30.00
Pot. Payout:
$139.92

Parlay (2 Picks)0 of 2 settled
Oct 27 • 2023 At 12:14:37 PM
Ticket ID: 6356561560
Moneyline: TOR Raptors @ +112
TOR Raptors @ CHI Bulls
Moneyline: GS Warriors @ +120
GS Warriors @ SAC Kings
Odds:
+366
Wager:
$30.00
Pot. Payout:
$139.92

[BCR]

Man, I’ve looked through these slates. I’ll bet 2 $30 parlays on Toronto-OKC and Knicks -GS

Those are +389 and +346. That’s the kind of return you need for this to work. I don’t dislike any of those teams shot at winning, but could see long stretches where you go 1 for 2 and lose. I think homecourt and crowds influencing refs might have more influence in NBA, plus load mgmt.

Those are more similar to what they did in those threads.

Personally for myself, I’m betting a SA-Den ML parlay, but +186 isn’t enough value for this experiment. Once you got 3 bets deep, even with martingale, you couldn’t get enough value.

Despite sweeping all my bets first round, what I did were more standard parlays. These 2 are more similar to what they did over on covers.

I’ll probably wait until MLB starts again. It is interesting though, and I liked reading the experiment. There is something there. I just doubt NBA is plausible.

Lines shifted a little by time I bet.


Parlay (2 Picks)0 of 2 settled
Oct 27 • 2023 At 12:14:11 PM
Ticket ID: 6356580747
Moneyline: NY Knicks @ -108
NY Knicks @ ATL Hawks
Moneyline: OKC Thunder @ +135
OKC Thunder @ CLE Cavaliers
Odds:
+354
Wager:
$30.00
Pot. Payout:
$139.92

Parlay (2 Picks)0 of 2 settled
Oct 27 • 2023 At 12:14:37 PM
Ticket ID: 6356561560
Moneyline: TOR Raptors @ +112
TOR Raptors @ CHI Bulls
Moneyline: GS Warriors @ +120
GS Warriors @ SAC Kings
Odds:
+366
Wager:
$30.00
Pot. Payout:
$139.92

The NBA man. I’m killing it so far with my bets, but literally every bet could win or lose with 3 minutes left.

Thankfully my personal parlay hit for +186. That’s worth 20x these other parlays. That took SA going to OT.

Both the small parlays were insane. OKC was down 10 with like 230 left and scored the last 13 points of the game.

But Toronto had a literal 99.9% win probability and lost in OT, and GS covered the back end. So 3 out of 4, but I won one that looked dead and lost one that seemed a lock win.

Either way. +$76.37 to the good on the NBA parlay experiment. +6.85 units over 2 series.

I guess I’ll continue another day despite it being irrelevant in the big picture. Still doubt it works long term despite its early success.

[aayjay]

Man, I’ve looked through these slates. I’ll bet 2 $30 parlays on Toronto-OKC and Knicks -GS

Those are +389 and +346. That’s the kind of return you need for this to work. I don’t dislike any of those teams shot at winning, but could see long stretches where you go 1 for 2 and lose. I think homecourt and crowds influencing refs might have more influence in NBA, plus load mgmt.

Those are more similar to what they did in those threads.

Personally for myself, I’m betting a SA-Den ML parlay, but +186 isn’t enough value for this experiment. Once you got 3 bets deep, even with martingale, you couldn’t get enough value.

Despite sweeping all my bets first round, what I did were more standard parlays. These 2 are more similar to what they did over on covers.

I’ll probably wait until MLB starts again. It is interesting though, and I liked reading the experiment. There is something there. I just doubt NBA is plausible.

Lines shifted a little by time I bet.


Parlay (2 Picks)0 of 2 settled
Oct 27 • 2023 At 12:14:11 PM
Ticket ID: 6356580747
Moneyline: NY Knicks @ -108
NY Knicks @ ATL Hawks
Moneyline: OKC Thunder @ +135
OKC Thunder @ CLE Cavaliers
Odds:
+354
Wager:
$30.00
Pot. Payout:
$139.92

Parlay (2 Picks)0 of 2 settled
Oct 27 • 2023 At 12:14:37 PM
Ticket ID: 6356561560
Moneyline: TOR Raptors @ +112
TOR Raptors @ CHI Bulls
Moneyline: GS Warriors @ +120
GS Warriors @ SAC Kings
Odds:
+366
Wager:
$30.00
Pot. Payout:
$139.92

The NBA man. I’m killing it so far with my bets, but literally every bet could win or lose with 3 minutes left.

Thankfully my personal parlay hit for +186. That’s worth 20x these other parlays. That took SA going to OT.

Both the small parlays were insane. OKC was down 10 with like 230 left and scored the last 13 points of the game.

But Toronto had a literal 99.9% win probability and lost in OT, and GS covered the back end. So 3 out of 4, but I won one that looked dead and lost one that seemed a lock win.

Either way. +$76.37 to the good on the NBA parlay experiment. +6.85 units over 2 series.

I guess I’ll continue another day despite it being irrelevant in the big picture. Still doubt it works long term despite its early success.

Nice, profits profit

[aayjay]

I Agree I think that baseball is best sport for this, at least best major 4. Wonder if it could be useful in college anything....

[PositiveVariance]

Very Interesting. 3 seasons worth of data is impressive, but I wonder if anyone has back tested a significant period of historical data (20 years or so)?

Also, I see flat betting came out profitable over the test period. If this is the case, It would appear to reduce variance by simply flat betting a higher base bet vs. Martingaling. The only way Martingaling would prove to be beneficial, is if it could be shown that "streaks" happen at a lower frequency with this strategy vs. selecting games at random, otherwise since these are independent events, martingaling would not have a greater expected value than flat betting alone.

I see the first post on this strategy was about 10 years ago. There must be some research as to why this may be potentially be +EV.

Also, when Martingaling 8 "levels", it can often times take awhile for the outcomes to resort back to the mean.

+Ev or not, very interesting!

[Sanlmar]

Very Interesting. 3 seasons worth of data is impressive, but I wonder if anyone has back tested a significant period of historical data (20 years or so)?

Also, I see flat betting came out profitable over the test period. If this is the case, It would appear to reduce variance by simply flat betting a higher base bet vs. Martingaling. The only way Martingaling would prove to be beneficial, is if it could be shown that "streaks" happen at a lower frequency with this strategy vs. selecting games at random, otherwise since these are independent events, martingaling would not have a greater expected value than flat betting alone.

I see the first post on this strategy was about 10 years ago. There must be some research as to why this may be potentially be +EV.

Also, when Martingaling 8 "levels", it can often times take awhile for the outcomes to resort back to the mean.

+Ev or not, very interesting!

3 seasons of Chrissy Martingale hasn’t affected your judgement?

[aayjay]

Very Interesting. 3 seasons worth of data is impressive, but I wonder if anyone has back tested a significant period of historical data (20 years or so)?

Also, I see flat betting came out profitable over the test period. If this is the case, It would appear to reduce variance by simply flat betting a higher base bet vs. Martingaling. The only way Martingaling would prove to be beneficial, is if it could be shown that "streaks" happen at a lower frequency with this strategy vs. selecting games at random, otherwise since these are independent events, martingaling would not have a greater expected value than flat betting alone.

I see the first post on this strategy was about 10 years ago. There must be some research as to why this may be potentially be +EV.

Also, when Martingaling 8 "levels", it can often times take awhile for the outcomes to resort back to the mean.

+Ev or not, very interesting!

Interesting indeed, some smart cookies involved in the discussion, bunch of morons tooo