axel

Most of us can't see all the matches played until a certain moment and make decisions for the next match based strictly on the past results and not on what really happened on the field.
It is impossible to fix it but we can try something.
Example:....
Round 17: C. Palace - Brighton 1-1
Shots on target: 4-10
Ball possession: 34-66
Don-t you think that C. Palace - Brighton 1-3
would have been a better result to be considered for the next decisions?
 

Another interesting picture:
I generated the ELO graph for top 5 teams in Premier, starting with 2005 season, almost 600 games.
1. There is a general downward trend, obviously.
2. Without underestimating the merits of Leicester, something happened to top5 in the 2016 season.
 

Probably most of us are wondering if the (successful) strategies available before the lockdown will work after the season is resumed   First of all,  my opinion is that during the lockdown was a bad idea to bet unless you have info about the teams involved in Belarusian second league. On other words, it would have been a kind of lotto for me but for a guy that lives in Belarus would have been a gold mine. As he knows the teams and the players will find easier the value especially because there will be a lot of guys like me that don-t even can read correctly the teams name.       Now, regarding the betting after the lockdown, I was curious what happened last 10 seasons considering the games where the home team have not played for at least 2 weeks. (there is no data for a larger interval between games) Because I have the data I calculated what happened if the interval between 2 games was under 3 games.   I noted T1 - top teams  ELO>2000 at the game's moment T2 - average teams  ELO>1700 at the game's moment T3 - bottom teams   Here is the graph only for "Home Win" market. How to read it, see the highlighted selection Please note that T1-T3 and T3-T1 had not so many games but for example, T1-T2 had a good sample.   (T1-T2)  means all the games between (Top teams - Average teams) played last 10 seasons   1. The first column (blue) - 74.47% of total games were won by the home teams   2. The second column(red) - 77.59% of the games were home team played the last games less than 3 days ago were won by the home team   3. The third column (green) - 55.56%  of the games were home team played the last games more than 15 days ago were won by the home team         As a first conclusion it-s easy to see that if the interval between games is  reduced, the top teams won more games than usual  but if the interval is larger  they won just a little bit over 50%   Now, just after the lockdown, we are in the column 3 situation so have to try to find value in laying top teams. After 1-2 rounds will be in the column 2 situation and for sure will try to back the top teams.   Of course, will need to check all the available info but I think this will be the trend.

Hi all,
in this thread I'll be sharing the results of my bets on ITF tournaments which I place using a model-based approach.
Why ITF?
ITF tournaments (lowest tier of professional tennis) seem to be an imperfect betting market where information is often scarce, so I assume bookies do not always know exactly what they're doing when they're offering odds. For Challenger/ATP/WTA tournaments my approach didn't work so well (albeit on a small sample) so I stopped betting there and concentrated on ITF. I bet on both men's and women's tournaments (no W60, W80, W100 because they attract a player field of a comparable level to ATP challengers which I find hugely unpredictable). The best bookies for ITF tournaments are bet365, Betclic. Bwin and Unibet occasionally also offer ok odds.
What does my model do?
It's a relatively simple Logistic Regression that predicts each player's win probabilities. Following main factors go into the model:
ELO rating (several versions calculated with different formulae - some place more weight in recent form, others are more influenced by long-term results) Recent changes in ELO rating Players' form on the surface played Head-to-head (overall/on surface played) Home advantage Additional factors I apply on top of the model as an "expert judgment" are:
Are "rising talents" involved in a match? (they tend to have ELO rankings lower than their true level) Are players from different playing fields involved? (In case 2 players meet, one of which mainly plays in Challengers and the other in Futures, the ELO of the Challenger player tends to underestimate his relative level because he plays in a stronger field; similarly, players from the Asian fields tend to be weaker than their European/American counterparts with the same ELO) Has the model performed well with the players I'm betting on/against? (if I bet against Roger Federer twice this week and he wins both times, I'm not likely to bet another time even if the model recommends me to) Do players have sufficient match form? (I won't bet on a player who hasn't played for 6 months if the model recommends me to) Tiredness (I won't bet on a player who yesterday played a 4-hour match). The probabilities generated by the model are then compared with to odds. I only bet on odds between 1.4 and 2.9. For higher odds, I found the model to be unreliable, for lower odds it just doesn't make sense as I live in Germany where all winnings on betting are taxed 5%. To account for model risks I apply an additional margin of conservatism to select my bets.
So if the model tells me that Player A wins with 60% probability, the lowest odds I want to take on Player A would be (1/0.6)*(1/0.95)*1,1=1.93. The (1/0.95) multiplier accounts for the betting tax, the 1.1 multiplier accounts for model error. In case I'm really convinced by the players' stats I might accept odds of 1.85-1.90.
The model is calibrated on an extensive history (ITF men's and women's matches starting 2001).
Final words
The 5% tax makes profitable betting in Germany pretty much impossible, so I only bet small amounts for fun. In this thread, I will show both profit before and after tax so people in other countries get a realistic view of the model performance. So far after ~950 bets I have achieved an ROI of around 7% before tax and 2% after tax which I find decent. I will always bet the same amount, let's say a symbolic 5 EUR on each match. Let's hope I can maintain my ROI after I start posting. I also hope this thread attracts some tennis betting enthusiasts, especially those who use statistical/ML methods and are willing to share their secrets because I truly believe a scientific approach can make betting profitable.

Hello, Hurco; welcome to the forum!
Your question is not silly; while they say that people are born with natural ability to swim, and then they lose that ability within first weeks of life, I highly doubt we are born with natural knowledge of betting exchanges, so some of us have to ask it sooner or later. I mean, all of us in this forum had such a question.
With your example, there is indeed not difference between exchange and "normal bet platform". However, betting exchnage offers you a possibility to trade your bets: to place the opposite bet before the event ended, so you lock in a profit - or, more usually, in a loss - regardless of final outcomeof the match.
Let's see how; in your example, you laid 0-0 at odds of 4.00 (insanely good odds for laying 0-0, if you ask me) with stake of 10 units, and liability of 30 units.
Odds for 0-0 tend to shorten during the match, untill they reach 1.01 in closing moment. After quarter of an hour of play, you're dissapointed with performance of your team, so you decide to trade out. By that time, odds have shortened to 3.50.
You place the opposite bet now: you back 0-0, with stake of 11.43. So, you're risking 11.43 units to win 28.58 units.
If they manage to find their way to net, you win 10 units from lay bet, and you lose 11.43 units from back bet; net loss 1.43 units. If, however, no goal is scored by the end of the match, you lose 30 units from lay bet, and you win 28.58 units from back bet; again, 1.42 units net loss. So, you were able to trade out when you saw that match does not develop as you expected, and instead of losing 30 units, you lost only 1.24 units.
Let's say that you laid the draw, at odds of 3.00, with stake of 10 units, so liability is 20 units. The goalkeeper has bad day, so there is a goal minutes into the match, and it sends draw odds to 8.00. Now, you back the draw, with stake of 3.75 units, for possible profit of 26.25 units.
If match does not end in draw, you win 10 units from lay bet, and lose 3.75 units from back bet; net profit 6.25 units. If bad keep team manages to claw back to the draw, you lose 20 units from lay bet, and you win 26.25 units from back bet; net profit again 6.25 units. Thus, you were able to lock in profit after first goal, regardless of how the match will develop further.
That example is known as "lay the draw" strategy; once profitable, it is worn out nowadays, and it is not easy to take profit from that.
This works the same way if you first place the back bet - you need subsequently to place lay bet, or you can trade more than once during the same match - place more than one back bet or lay bet at different odds; if back odds are higher than lay odds, you make profit, otherwise, you make loss.
Stakes above are calculated using formula:
Back stake * Back odds = Lay stake * Lay odds
and it does not incorporate comission at betting exchange, you need to deduct it from your winnings; Betfair has 7% basic commission.
Now, bookmakers started offering "cash out" option, which is very similar to this trading, as long as your bet still has the ability to win. Once it is losing, of course they won't offer you possibility to decrease your losing stake.
Hope it was understandable... Good luck with your trading!

Your observation is totally correct, you are absolutely right!
You only need to pay a bit of attention to your words that I emphasized above. If you check, you will see for yourself that odds for under 5.5 if one goal is scored at HT, and under 4.5 if no goals at HT, are barely 1.05 or even less. And if you check this year games, not a dozen of today games, but thousands of this year games, you will find that usually 5 out of 100 such a matches go over 5.5 or over 4.5. So, if you bet 100 units each time, you will win 5 units 95 times, add or take two or three times, and lose 100 units 5 times, add or take two or three times. Now do the math, and you will see where does your system go.
Though, if you are able to select games, not to bet on all today games or all this year games, but to select matches so that you hot over 4.5 or over 5.5 less than (100-100/odds) out of 100 matches, where "odds" are average winning odds you get, after commission, then indeed you have a profitable system. So, all you need is to find such a matches.

to Punters Lounge @Willygib  

A few days ago I had a discussion with some friends about two very controversial topics.
I opened the discussion here because both topics are the basis for many betting strategies.
It's about "gamblers fallacy" and "value in betting"
 
There are probably just a few who don't know what we are talking about here, so I will remind them:
Value, in betting, appears when the odds are suggesting that the chances to happen the event are smaller than they really are.
And
according to Wikipedia: The gambler's fallacy, also known as the Monte Carlo fallacy or the fallacy of the maturity of chances, is the mistaken belief that if something happens more frequently than normal during a given period, it will happen less frequently in the future (or vice versa).
Now, the problems in discussion were, and I would appreciate your opinion
1.
We talk every day about value but is it possible to calculate the exact value for a price so we can tell if we will have profit on long run?
If we have the price 2.10 for 50% event like coin flip  it is a clear value and will have profit on long run  but,
For example we have 1.90 odds for home team win and we know with 15 minutes earlier than anyone else (my brother’s best friend is  brother in law with the wife-s manager) that the away team will play with the second team or something similar.  
On this particular situation the odds 1.90 is a certain value and should hurry to get it?  Why? Common sense tell us that should be value, but based on what? We bet based on common sense or based on value?  What is the probability for home team to win event? What way is calculated?
I agree that there are way more chances to have value but my opinion is that will never know if we bet with value or not. Until will see that checking the money into accounts of course.
2.
I have often warned about the danger of “gambler’s fallacy” but I was right regarding to betting on football?
We have an average team that won 5 in a row. Would be correct to start a strategy like Martingale against them?  At first sight would be terrible wrong if looking to “gambler’s fallacy” concept.
But, extremely important, the concept is about independent events. If there will be red 100000 times the next red/black will have the same 50% probability to appear.
This is the same thing for our team?  The next game is not affected at all by the previous games? Some players are probably getting more and more tired.  And many other similar things that affects next games.  
I am against Martingale in betting but my opinion is that is am extremely good strategy in some very particular situations.
 
 

My opinion is that sport betting is learnable for sure but the problem is that you will need more and more resources (time for instance) as you are closer to real profit. Not everyone has patience and resources to do that. (I have not )
It is relatively easy and with a minor effort to study some afticles, books..anyone can transform betting from a black hole into a thing that will not give money but will not suck your money neither.  Maybe will just pay the correct price for the "fun" you have when your bet is win/lost in 92 min.
If you want to win money from betting, first off all you need to check if you "deserve" it.  Did you forget about luck/badluck and accept that s about knowledge and hard workl? Do you know more than other 99% of the bettors the teams involved in your selections? Are you able to calculate in short time complex probabilities?...and maybe there are another 10 similar questions. If you have just a single "NO" then is not good and... back to school.
Regarding the previous comments, probably what posted notanotherdonkey is not related to betting but it brings to attention a very impotant problem the begginers have to learn: the gambler's fallacy concept.  On short, from Wikipedia : " is the mistaken belief that if something happens more frequently than normal during a given period, it will happen less frequently in the future (or vice versa). "
 
 
.     
 

I hope we all agree that the value can be found in very small odds but, as in real life, to study the small things you need fine tools and a lot of patience, otherwise you destroy them. It is best to avoid small odds, in my opinion. Is best to avoid very "crowded places" as well. No value where are too many players.
My first impression, as did my colleagues above, was that such a strategy simply applied cannot be mathematically winning.
 
I checked all the matches in September considering the difference at least 10 places for the league positions, see picture no.1.
157 out of 1635 met the condition and, as expected the results would have been not extraordinary as you can see in picture no.2


 
Now, lets-s see what if will consider the places in the home table for home teams and in the away table for the away teams. We have some improvements already. Almost 7% profit for home win. It makes sense because we filtered teams with a good play at home vs teams with a bad play away (and too many players consider only overall teams performance)
See pictures no.3 and no.4
 


 
Now, let-s go a little bit further and check for  home teams placed between 1-5, so they are very strong at home, and the odds over 2.00 (the potential has not been seen yet by the majority of the payers). 
See pictures no.5 and no.6


 
The results are much better. See picture no 7.

 
Of course, the results are based on aveage odds but I am sure you got the idea. Don-t play what another 99% play becasue it-s not possible to have 99% of players winners on long term.
 
Finally,
of course, I will not be a hypocrite and to say that I did not want to show off the possibilities of my program. 
 

I think it is a very important topic.
Until one of the experienced guys will help us with the answer I will say my opinion.
 
I think that one of the mistake the player does is to increase the stake too early.  We have to understand that betting is a game of probabilities and the probability to lose all bets for the rest of our life is not 0. It is extremely small but is not 0. Of course this is an extreme example..
The most logical answer is to increase to 2 units when your bank is 200, in order to keep 1% from bank. Probably this would say most of the players.
The second logical answer would be to never increase the stake,  and to save every 100 you get. Probably this would say just a few.
 
I prefer a combination: I would keep the stake 1 until I have 300, then save 100 and increase the stake to 2. 
Now I have 200 and stake 2, still 1% but have 100 saved (my real pofit). A few comments here: Another mistake can be done is to think that the balance you see on your bookie, under your account are yours. No, the money are yours only when save them to your bank account.
 
Will keep it until I have 400, then save 100 and increase to 3...
And so on  
 
As I said, this is what I think in this Monday early morning, the worst hours for my brain in the entire week

to PL @axel   

Racing & Football Outlook - it's a UK based paper focused on, you guessed it, Racing & Football but more betting orientated.