Quote from: Bardman on September 28, 2016, 11:41:27 AM

Quote from: PrediMarket on September 27, 2016, 04:07:26 PM

Quote from: Bardman on September 27, 2016, 03:00:03 PM

Quote from: PrediMarket on September 27, 2016, 12:33:59 PM

Quote from: Bardman on September 27, 2016, 11:55:42 AM

It seems very hard to find two different bitcoin websites that will give you high enough odds to make that work. I didn't really understand the second part and your formulas are really not clean, you could have made it much simpler.

For fiat money websites, you do have comparison tools which show you sure bet opportunities like http://www.bmbets.com/sure-bets/ (not my site fyi).
For bitcoin betting sites, there might be similar tools but none that I am aware of.

In order to find such opportunities, your best chance is to hover websites which have low overround (i.e. margin) like https://predimarket.net (this is my site fyi) and make a little spreadsheet with predefined formulas computing the overround.

And what is the second part you refer to?

The part with the bookie and their odds. So you are basically calculating the profit based on what the bookie expectations or odds are but how can you take advantage of that?

I assume you are talking about "value betting" not "sure betting".

A value bet is a betting opportunity where you think one outcome (e.g. the victory of whatever team or player) is more likely to occur than the odds of the bookmaker actually suggest. If you check the calculations in the original post, I am showing how to compute the "opinion" of the bookmaker towards the game. Using the formula I gave, you can calculate the probability of victory of team A according to the bookmaker and see what he actually thinks is going to happen in the end.
I also gave a formula to compute the margin that the bookie takes, to give you a better idea of how "greedy" he is.
The opinion (i.e. probabilities) and margin are simple metrics that can help you decide if the bet is worth something or not.

As per how to take advantage of that, you need to first "decide" what your own probabilities are and then compare them to the odds of the bookmaker (see OP).

An example of that:

A plays against B and you can bet on either A or B.
Assume that you compute the "opinion" of the bookmaker and find that he thinks that A has 60% chance to win and B has 40% chance to win.
You also compute the margin and find 5% (that makes an overround of 1.05).
(All of this combined means the odds for A and B are respectively 1.587 and 2.380)

Now you need to "decide" what is your opinion of the game A vs B, in terms of probabilities: is it 65% for A and 35% for B, or maybe 70% | 30% or even 60% | 30% like the bookmaker. These values are yours to decide, they are up to your knowledge and expertise of the game.

If you think A has 65% chance of winning, then you compare this value to the inverse of the odds for A i.e. 1/1.587 which is equal to 0.630 roughly: you can see that it is smaller than 65% (=0.65). In these conditions, betting on A is an opportunity from your point of view and you should bet on A.
On the contrary, if you think A has no more than 60% chance of winning, then you should not bet on A (at this bookie at least) given that 0.60 < 0.630.

Of course, there's a thin line between assessing the chances of winning at 60% or 65% but this is what can give you an edge. And more realistically, you should only bet when you notice a significant difference between your probability and the bookmaker's, just to give you room for estimation errors.

Ok but in that case couldn't you profit just from copying the bookies predictions? Always betting on the team he thinks has better odds, obviously the profit won't be much but as you said they are pretty accurate, not 100% so you might as well just copy them no? After all when you bet on a team you always bet on it because you think its gonna win no matter what the bookie says, have you tried your "strategy" long term?

If you share exactly the same opinion as the bookmaker, then making the bet with this bookmaker is always a bad idea.

Take my last example: if you think A has also 60% chance of winning, then you compare 60% (=0.6) to 1/1.587 (i.e one divided by the odds of A) which is equal to 0.630 roughly.
Given that 0.6 is smaller than 0.630, then betting on this game is a bad idea (at this bookie at least).

It is actually important to consider the differences between your opinion and the one of the bookmaker. Because then, you can apply this rule (if you choose to go for value betting) and in the long run (after a significant number of bets), that kind of betting discipline will be fruitful.

About your question regarding myself using those strategies, the answer is no and here's why.

Firstly, I wanted to make my own sportsbook (PrediMarket) because I have been developing a whole range of pricing techniques for bookmakers that I wanted to put into practice.

Then, sportsbetting being a very competitive field, I thought I could also turn myself as well into a gambler, to increase my revenue but I would only do that if I could automate a betting strategy because it is the accumulation of bets that can yield a relatively sure profit.

However, for the two basic strategies I described, there are significant obstacles that I cannot overcome for now:

sure betting strategy: you need at least 2 bitcoin bookies which provide an API but currently, there's only one that I know of (BETBTC)
value betting strategy: I would need to create an algorithm to produce probabilities for each game, based on historical data. While I have already done that for simple cases (like intra-league games), it gets more complicated when you are dealing with games involving from different leagues like UEFA Champion's League which might have never played against each other.

Although, it is difficult to automate such strategies, you can still apply them at a human level and they will prove useful.
If I tell you I hold a PhD in Applied Mathematics in Economy, would that convince you a bit more?