Arbusers

A week ago, I started getting valuable betting odds on orbit, pinaccle and a local Turkish site. I think I have enough statistical data. You can see the screenshot of my excel table in the attachment. I am well behind the expected value.

please advise me.
1.Do you think this deviation is normal?
2.Should I change my valuebet source?
3.Should I keep betting?

Firstly, I would like to thank you for both valuable posts you made today, I really appreciate it.

A couple of questions:
1. How do you know the bets you have contain value?
2. How do you calculate the expected value? Based on what?

You mentioned Orbit and Pinnacle. I find it extremely difficult to find value there, even though I m not saying this is not possible. I am noticing from the screenshot that you use PS3838, so I am wondering what is the name of the Turkish site?

Especially for Pinnacle, recently a member of our forum sent me a video of a Pinnacle figure that worths watching. In this video he describes how pinnacle finds value and what is the mindset of their traders. Have no doubt, pinnacle can be an opponent and an ally at the same time. But knowing the way an opponent thinks or acts would add more value to your way of thinking.





https://www.youtube.com/watch?v=8VpaC-4qYQY


I understand that your betting sample is already nearing 300 bets, so I consider it adequate to determine if you are doing something good or bad. I believe you should drop it. Unless there is a monstrous deviation from the average, you can't expect good things to happen with this tactic.

Thank you too for your reply.

1. The valuation is offered by a surebet provider that is not on your site. I was using it for surebet before, but as you can guess, many soft bookies limited me and I just started trying Valubet.

I doubt they are offering wrong values. I'm just copying the winning probability and overvalue from that site. I created the formulas for the remaining calculations myself.

I'm betting up to odds of 3.00 max and I have fixed my max return amount of 300 euros including my stake. Although many markets do not allow me to go to these limits, this is my basis. Generally, my low stakes won and my high stakes lost. maybe this is where the problem comes from. I also placed bets with at least 2% overvalue.

2. Slots usually have a house edge of 3.5%, which means 3.5 euro profit for every 100 euro game for the casino owner. From the player's point of view, a loss of -3.5 euros is expected for every 100 euro game. For this reason, I thought like a casino owner about the expected value in order to be able to project forward, so this value depends on the amount I bet (column named stake in the excel table)

Let me explain with an example.
probability of winning 50%
ps3838 odds 2.20
overvalue 10%
stake $100
The expected value will be $10 here.

The name of the Turkish site is holiganbet.

In May, I made a trial with 111 bets. The results were positive at 769 euros (expected value 453 euros). I have added an image about it. That's why I decided to try it extensively in June.

Sorry, your video is not showing. If you share it in the form of a link, I would like to watch it, I would be glad.

Also, what I'm curious about is how many % deviations can be experienced depending on the wager only for those who make valuebet.

Put all this bets in an excel. Make a normal distribution, and calculate the probability of being over zero profit. In my opinion, you are not catching value. I am not saying that there are no bets with value. It means that all the bets together don't have value. You made more than 330 bets and the odds are less than 2. And i suposse that your stakes were "homogeneous", So, with a simple view, and without doing the maths here, i think that you are close to the "expected hope" close to negative 6%.

Which Alert Service are you using here?

From what you say I understand there might be a problem in your staking plan too. What is the reason for cutting the underdogs with odds above 3.00? Some times the staking plan might be of equal or bigger importance than the overall plan.

Here is another question: How is this surebet provider determines value? Is it only because the odds are simply... arbs? Do we call that value? Did they explain one way or another. Also, are we talking about pregame bets or in play?

CharlieSheen99's advise is good, and I would take it one step further by creating various filters that would allow you to feed it with information and then have the meta-analysis that would show you the mistakes. For example, what if the analysis shows you that you are bleeding money with Orbit, and you are gaining money from Holiganbet?

ozcelik wrote: ↑

Thu Jun 08, 2023 3:13 pm

Sorry, your video is not showing. If you share it in the form of a link, I would like to watch it, I would be glad.

I placed a direct link to Youtube right under the video in my previous post. Hope that works. Maybe your government is blocking some social media content?

I found the source of the problem yesterday. The wrong choices I made in the provider's filters drove me to the bottom.

The provider scans the odds of more than 450 bookmakers (included clons) and detects valuebets by comparing 2 to 50 of them. My filter was always set to 2. I thought this represented 2-probability outcomes.

The problem is that if I choose 2 in this setting, 30 valuebet comes out. But if I choose 50, the number of valuebet drops to around 10. In other words, 1/3 of the bets I made were wrong. In the basic comparison, 2 bookmakers are insufficient and this caused me to dive into the margin of bookmakers.

However, I wouldn't expect 7% loss in a bookmaker with a very low margin like pinaccle. Because their average margin is 3%.

For this, I deepened my analysis in my excel table yesterday and saw that I lost 1000 euros each on hooligan and ps3838. orbit was 36 euros in profit. So ps3838 got the "expected hope" from me, around 3%.

arbusers wrote: ↑

Fri Jun 09, 2023 7:14 am

From what you say I understand there might be a problem in your staking plan too. What is the reason for cutting the underdogs with odds above 3.00? Some times the staking plan might be of equal or bigger importance than the overall plan.

I did this to reduce the risk of losing the bankroll completely. When betting 100 euros at odds of 3, 200 euros at 1.5 odds and 150 euros at 2.00 odds seemed reasonable. I thought that with this tactic I could tolerate up to -3000 euros in the worst variance. If you have a recommended odds range and staking plan, I'd love to hear it.

arbusers wrote: ↑

Fri Jun 09, 2023 7:14 am

I placed a direct link to Youtube right under the video in my previous post. Hope that works. Maybe your government is blocking some social media content?

Your video is now visible. Thank you. Yesterday there was only a dark screen.

I am new to Valubet, so I would like to hear all your advice and suggestions on this subject.

Hi Ozcelik.

Nice to meet you. I'm also new to the forum and trying to read as much things as possible in order to come up with my settings for valuebetting.

It seems that we are in a similiar position. I'm also in a country in EU with a lot of betting restrictions therefore I have to use BetinAsia.

At the moment i'm using Molly (Black BetinAsia) and PS3838 for my valuebetting but my settings generate a really low number of bets per day. I get my valuebets from the same free source as you do:

Filter 1:
[*] Event During: 3 Hours Max
[*] Minimum Odds for Calculation: 15
[*] Odds: 1.3 - 2.5
[*] Overvalue: +5%

Filter 2:
[*] Event During: from 3 hours to 96 hours
[*] Minimum Odds for Calculation: 15
[*] Odds: 1.3 - 2.5
[*] Overvalue: +8%

I still don't have enough data to be able to share with you but i have a few questions that might address some of the issues you are facing.

1. Have you tried to calculate the same bets using the kelly criterion? Yesterday i did it for a database that I have of 400 bets (PS3838 and 22bet) and the results were much better than my fixed staking plan.

2. I have seen that you covered the issues on the "number of bookmakers compared". I have it for 15 and will continue doing so to see what it happens in the long term.

3. Try increasing the Overvalue %. Of course you will have less bets but definitely increase ROI.

4. How many hours before in play do you have on your filters?

Is there a private chat were we can talk? We are in a very similiar position regarding the software used and bookmakers.

Cheers.

I was not using fixed stakes. my max stake: stake+profit=300€. Due to pinaccle limits, I couldn't bet much on this limit.

I compared my bankroll management strategy with the kelly criteria. I calculated according to 30% kelly recommended by value bettors. and %100 kelly.

average stake (total wager/number of bets)
me:93€
30% kelly:47,19€
100% kelly:41.28

expected profit
me: 4.18%
30% kelly: 6.34%
100% kelly:6.48%

current profit
mine: -6.27%
30% kelly: -10.35%
100% kelly:-13,92%

It would be wise to apply 30% kelly because I would lose 20% less than my current loss. and the increase in changing expected profits is a good reason to comply with the kelly criterion.

100% kelly lost 1.5% less than me and that's only in 300 bets.

I added the graphics.

2. The safest choice out there would be 50 bookies but unfortunately that is not enough. The numbers they offer as the probability of winning do not reflect the truth. just today i compared their valuebet with oddsportal. They showed 9% profit in the basketball over-under market with 7% house edges. so in fact, the actual value could have been at most 2%. (ignoring the closing line value) You'll see inconsistencies like this when you compare.

3. I've already done that. I was betting on events with a maximum odds of 3.00, initially offering 2% value and starting in 2 days. I then set the value to 3%, then to %4, etc. I finally reached the expected value of 4.18% on average. but since the probability percentage they gave was wrong, I always fell below the expected value.

4. I have set it to 12 hours or 8 hours since I opened this topic. before that i was setting it to 2 days.

The settings don't matter. they are not focused on valuebet, arbitrage is their business yes. they come to a conclusion based only on the average market rates and adding the margin part, which is unknown how they are calculated. because there are different margins for each market. I know sites that have a 12% profit margin on player bets and people are betting there like crazy. I think there is an error in margin calculations. The best method is not to bet on any valuebet below the 7% they show (as the pinaccle margin is up to 6% in small markets). this applies to pinaccle, but here you face another problem; valuebet bots. While you have just opened the event page, they have placed their bets for seconds.The odds have dropped suddenly or the market has suspended.

I've decided not to bet on their value anymore. I would not advise you to trust them.

And have you tried playing the opposite side?

ozcelik wrote: ↑

Tue Jun 20, 2023 1:25 am

I was not using fixed stakes. my max stake: stake+profit=300€. Due to pinaccle limits, I couldn't bet much on this limit.

I compared my bankroll management strategy with the kelly criteria. I calculated according to 30% kelly recommended by value bettors. and %100 kelly.

average stake (total wager/number of bets)
me:93€
30% kelly:47,19€
100% kelly:41.28

expected profit
me: 4.18%
30% kelly: 6.34%
100% kelly:6.48%

current profit
mine: -6.27%
30% kelly: -10.35%
100% kelly:-13,92%

It would be wise to apply 30% kelly because I would lose 20% less than my current loss. and the increase in changing expected profits is a good reason to comply with the kelly criterion.

100% kelly lost 1.5% less than me and that's only in 300 bets.

I added the graphics.

2. The safest choice out there would be 50 bookies but unfortunately that is not enough. The numbers they offer as the probability of winning do not reflect the truth. just today i compared their valuebet with oddsportal. They showed 9% profit in the basketball over-under market with 7% house edges. so in fact, the actual value could have been at most 2%. (ignoring the closing line value) You'll see inconsistencies like this when you compare.

3. I've already done that. I was betting on events with a maximum odds of 3.00, initially offering 2% value and starting in 2 days. I then set the value to 3%, then to %4, etc. I finally reached the expected value of 4.18% on average. but since the probability percentage they gave was wrong, I always fell below the expected value.

4. I have set it to 12 hours or 8 hours since I opened this topic. before that i was setting it to 2 days.

The settings don't matter. they are not focused on valuebet, arbitrage is their business yes. they come to a conclusion based only on the average market rates and adding the margin part, which is unknown how they are calculated. because there are different margins for each market. I know sites that have a 12% profit margin on player bets and people are betting there like crazy. I think there is an error in margin calculations. The best method is not to bet on any valuebet below the 7% they show (as the pinaccle margin is up to 6% in small markets). this applies to pinaccle, but here you face another problem; valuebet bots. While you have just opened the event page, they have placed their bets for seconds.The odds have dropped suddenly or the market has suspended.

I've decided not to bet on their value anymore. I would not advise you to trust them.

I thought my telegram address appeared on my profile. We can talk there or I don't know if there is a private message here, it can be that way.

Hey man,

Thanks for the answer.

Maybe you are right and they are not a very reliable value bet finder plaform. I've done almost 100bets at PS3838 and these are my stats at the moment (see attached).

Consider 30% Kelly.

Average Overvalue % is really good, but it definitely takes so many bets out of the equation.

What do you think of it?

We do not support the exchange of Point of Contact. Please refrain.

arbusers wrote: ↑

Tue Jun 20, 2023 3:36 pm

We do not support the exchange of Point of Contact. Please refrain.

I'm sorry. Will not repeat that in the future.

Cufs wrote: ↑

Tue Jun 20, 2023 3:08 pm

ozcelik wrote: ↑

Tue Jun 20, 2023 1:25 am

I was not using fixed stakes. my max stake: stake+profit=300€. Due to pinaccle limits, I couldn't bet much on this limit.

I compared my bankroll management strategy with the kelly criteria. I calculated according to 30% kelly recommended by value bettors. and %100 kelly.

average stake (total wager/number of bets)
me:93€
30% kelly:47,19€
100% kelly:41.28

expected profit
me: 4.18%
30% kelly: 6.34%
100% kelly:6.48%

current profit
mine: -6.27%
30% kelly: -10.35%
100% kelly:-13,92%

It would be wise to apply 30% kelly because I would lose 20% less than my current loss. and the increase in changing expected profits is a good reason to comply with the kelly criterion.

100% kelly lost 1.5% less than me and that's only in 300 bets.

I added the graphics.

2. The safest choice out there would be 50 bookies but unfortunately that is not enough. The numbers they offer as the probability of winning do not reflect the truth. just today i compared their valuebet with oddsportal. They showed 9% profit in the basketball over-under market with 7% house edges. so in fact, the actual value could have been at most 2%. (ignoring the closing line value) You'll see inconsistencies like this when you compare.

3. I've already done that. I was betting on events with a maximum odds of 3.00, initially offering 2% value and starting in 2 days. I then set the value to 3%, then to %4, etc. I finally reached the expected value of 4.18% on average. but since the probability percentage they gave was wrong, I always fell below the expected value.

4. I have set it to 12 hours or 8 hours since I opened this topic. before that i was setting it to 2 days.

The settings don't matter. they are not focused on valuebet, arbitrage is their business yes. they come to a conclusion based only on the average market rates and adding the margin part, which is unknown how they are calculated. because there are different margins for each market. I know sites that have a 12% profit margin on player bets and people are betting there like crazy. I think there is an error in margin calculations. The best method is not to bet on any valuebet below the 7% they show (as the pinaccle margin is up to 6% in small markets). this applies to pinaccle, but here you face another problem; valuebet bots. While you have just opened the event page, they have placed their bets for seconds.The odds have dropped suddenly or the market has suspended.

I've decided not to bet on their value anymore. I would not advise you to trust them.

I thought my telegram address appeared on my profile. We can talk there or I don't know if there is a private message here, it can be that way.

Hey man,

Thanks for the answer.

Maybe you are right and they are not a very reliable value bet finder plaform. I've done almost 100bets at PS3838 and these are my stats at the moment (see attached).

Consider 30% Kelly.

Average Overvalue % is really good, but it definitely takes so many bets out of the equation.

What do you think of it?

I don't think 100 bets is enough for forward projection. my data was based on 960 betting results.

your current position may be the result of positive variance that you appear 10% above the expected value. In addition, we need to refer to the closing line so that we can determine the actual expected value.

still, 15% expected value is a very good value and even if 7% of it is bookmaker margin and 5% is below the closing line, it may be enough to take you to 3% profit. these are just my guesses, it's hard to say for sure because they are offering the wrong value on that site.

Thanks for the answer's man.

I don't know if I will be able to get to 1k bets for this system but maybe would allow us to understand if the valuebet finder works (even if he does not know how) and if these filters are good.

I have read somethings about valuebetting in sharp bookmakers and maybe there's not enough value to do it only with the support of a valuebet finder software. These softwares get their valuebets odds from the correlation between true odds from sharps and bad odds from soft bookies. Therefore if we are using sharp books to bet it will be hard to get value. Do you agree on this?

I guess we should add something more to the system other than a basic filter of hours before in-play / number of bookies compared / odds placed / etc. Might be where the holy grail is.

A few days ago I had an occurance, but used no tool to compare it:
[*]Got a Overvalue Bet of 38% on Venus Williams vs Camila Gorgi (1set winning by Camila)
[*]I thought that something was wrong about it and my POV is that Pinnacle (PS3838) was the first to notice something 1hour prior to in-play. They corrected the Odds (going up from 1.55 to 1.95, something close to this if i'm not wrong). So in comparison with the other bookies, if Pinnacle was the first to do it then the valuebetting software thought this would be a valuebet. But of course it was not! There was no value, just a late update from soft bookies vs Pinnacle.

Can you tell me what do you do to counteract these possibilities?

ozcelik wrote: ↑

Tue Jun 20, 2023 1:25 am

I was not using fixed stakes. my max stake: stake+profit=300€. ... I was betting on events with a maximum odds of 3.00 ...

Thank you so much for the sharing of data Ozcelik.

I am hijacking this thread (sorry) to comment on those two points:
1. fixed stake vs variable stake
2. preferable odd ranges.

I have of thought about those for a while, but only today I have looked into this more in detail.

So I have been value betting for a while. In recent years mostly just shop betting. During every bet I have tried to find reasons to downsize my standard stake if there are some uncertainties in order to minimize risk. In some occasions I would also amplify the stake, but that does not happen so often and often it is not possible as shops have limits as well.

So how have I been performing? Well, it seems that my efforts have been in vain. After almost 15000 value bets I am sitting at exactly the same yield with a variable stake, than if I had used an imaginary fixed stake. Of course, this approach has helped me reduce variance during tough times as I would always stake less on odds 4.0 (for instance 50%) than on odds 2.0 (100%). But suppose that the bank roll gets sufficiently large so that all that matters are limits. Losing streaks may be tougher but that is more than compensated by better winning streaks.

With regards to preferable odd ranges, I used to go mainly with 1.5 to 3.0 in order to minimize variance, maybe 10% were above that. There seems to be so much value left on the table avoiding the higher ranges, so in order to maximize my time in the shops I now take many single bets on higher odds as well. My highest yields are on the higher odds, they are terrific. One reason could be that the soft books I use have odds models that seem to work very poorly in the extremes, such as when one team is a huge underdog. So the new saying is: bet to the max! it gives you more reason to spend your time doing it.

Value bets in the lower ranges below 1.5 are also included and parlayed with other lower odd value bets to amplify value. I am a bit more careful about those though. The reason for this is that I suppose a small handful of the events that I bet on are fixed or strange matches where regular assumptions on value don't apply. Odds do not reflect probability at all. That does not matter much for the very high odds where few tickets are won, but for those very low odds that are expected to win, that can be a problem I think.

I have attached some tables that I use to make assumptions upon. Yields are quite high in general and they have been rising through the years. I think one of the reasons is effective filtering, assuming which markets to avoid (or play them differently than the rest). Also, the books I use are quite weak in getting the odds right, yet great in expanding and creating addictive machines. So getting high yields on those books is easy compared to some of the better softs, which again would be easy compared to the sharps. There is a sort skill hierarchy among all books.

If anyone have experiences regarding "fixed stake vs variable stake" and "preferred odds range", please share.