Personal complaint lol

[Bayes]

I  am registered but when I  am trying to place a bet the site turns  into a green background.

I can't even register. I just get "loading" in the top left corner of the screen, but nothing ever loads.

[TurboGenius]

You have to factor in the house edge when ranking players - not how much they bet.
Say player A) uses 100.00 chips on straight up numbers but player B) uses 1.00 chips
Player A will have placed 100 times in bets more than player B even though player B could be (should be) leading in the ranks.

This is because the amount being bet doesn't matter (or shouldn't in the ranking really).

Euro wheel -
Each cycle of 37 spins would mean the house edge is 1 unit (unit being the average bet per spin)
Then you can figure out who's doing better than someone else by how far ahead of this expected negative % that each player is.
(yikes, confusing my own self lol)

1.   FreeRoulette   304400   1129   3627685   3898880   1.0747570

Ok, for example I'll use this stat from the leaderboard page.
$3,898,880.00 bet over 1,129 spins. This means $3,453.39 being bet per spin average (how that's possible, I have no idea lol)
So with 1,129 spins, we have 30.51 cycles of 37 spins.
The expected math result (what the player should have considering the house edge) is -$105,362.93
(amount bet average per spin X cycles of 37 spins)
The player has (won) $3,898,880.00 - (bet) $3,627,685.00 or $271,195.00 profit
Then you compare this to the expected amount and represent that as a percentage.
This percentage is how you can compare members and place the ranking.
Someone might be well above the math-expected amount but ranked low as it currently stands.
When you compare this way, it doesn't matter how much a person bets - small units or large units -
the percentage is what matters. (the player's 'edge' if there is one)
This net % is the perfect way to see how players rank because it's not relevant what unit size they use, like it is now
in the current ranking. Betting 1.00 chips and winning vs betting 100.00 chips and winning - you have to use this % to compare them or else there's no use in ranking lol.

Below is what it looks like when a person bets 1.00 on 1 number for 37 spins and everything works normally (as expected)
They have 1 winning spin and 36 losing spins - the house edge that was obtained is obvious.
(note the -2.70% just as it should be)


here (pulling from my own sessions not done on the multiplayer site) :


So using this value to rank is perfect, it won't matter what amount someone bets, but what will matter is what everyone is looking for - how to beat the house math edge and by how much.

Just my 2 cents. the "winrate" column needs a rework IMO

[TurboGenius]

Everyone "should" - given enough time - have a -2.70% in the reworked "winrate) stat column. (note "enough time" could be an eternity or it could be 10 spins - it depends on if they are winning and have something that works or not.)
Then you can look at players that are positive in this column and compare that to people at -2.70 where the math says they should be at... and others with worse than -2.70 who are actually losing worse than they should.

and now.... I must rest..... LOL

[Bayes]

I agree with Turbo. It's a simple and fair ranking system. The formula is -

(units Won - units Lost) x 100 / Total units bet = Edge (as a %)

Of course some will get lucky for a while, but more spins will sort out the men from the boys.

[Steve]

I don't know how the rating system works for the game - is it posted somewhere?

It's:

Your rank score = A x B x C
                         = (your amount of spins played / AVG of all players) x (your amount bet / AVG of all players) x (your win rate)

Where A is never greater than 1, and B is never greater than 1.

And your "win rate" is (amount won) / (amount lost). So a break even win rate is 1.000000. We could put it in terms of edge, but many people dont even understand what -2.7% would mean. And besides it wont change who's ranking higher if win rate was calculated the same way as "edge"

[Steve]

I  am registered but when I  am trying to place a bet the site turns  into a green background.

If you have problems, it would be something like browser security settings. I've tested it on ipad, android, windowws 7, 8 and and 10 with no problems.

What`s the idea for creating multi player feature anyway?

For fun, chat, separating people who are all talk from people with a good system, and competition. It isnt possible with dublinbet.

[Steve]

Turbo what you described is much the same as the "win rate". But the "win rate" alone is not a reliable indicator of success, which is why its easy to get a 2.0 win rate in short term, and hard to get a 1.1 win rate in the longer term.

The way the current leaderboard reads is players who have the highest win rate AND have played comparatively more than others are ranked at the top. The only way you can have a high win rate and rank at the top is if you have played enough spins and made enough bets to back it up.

So unless I'm missing something, what you described as your suggest algorithm is already the existing win rate. A win rate of 1.2 is the same as 120%.

Also betting $3000 or is easily possible. just max bets on 3 outside areas and a few inside bets.

The data from roulette xtreme is mostly irrelevant, like highs, lows drawdowns etc. Again what matters is amount of bets placed, amount of spins played, amount won and amount lost. The win rate is simply (amount won) / (amount lost). We added the variables or amount of spins and amount bet to ensure that "short term lucky" players dont rank high. The algorithm is quite simple, but is working as intended. The intention is again players who have the highest win rate AND have played comparatively more than others are ranked at the top.

It does not help to have a running bankroll because one player may have -$100,000 and another -$10,000 but they both have played just 100 spins. What matters is amount won vs amount lost and how much they've played.

Please explain what I'm missing. What you are basically proposing is to rank players based on edge alone. This is the same as ranking on win rate alone. But they how do you propose we deal with short term players with high win rates?

[Steve]

(units Won - units Lost) x 100 / Total units bet = Edge (as a %)

"Win rate" is (amount won) / (amount lost). And we already factor in total units bet based on 1 unit = $1. A player can instantly change the value of their unit (hypothetically), so it needs to be consistent for everyone at $1 units. What happens when the player bets $5000 on ONE spin and wins? They just got very lucky and will rank very high. They could even stop playing and out-rank players that have a 1.1 win rate over 1,000,000 spins and $10m bet.

That's why we need "amount of spins" in the equation too.

It's true that one player may play with $1 as their units and another may play with $100 per units. But I dont see a way we can deal with this without creating additional problems. The same table limits apply to everyone, and the goal of roulette is to win the highest amount of money possible. So everyone is on a level playing field. If a player wants to make lower bets, thats their choice. But to outrank others they'll need to make larger bets.

[TurboGenius]

Your rank score = A x B x C
                         = (your amount of spins played / AVG of all players) x (your amount bet / AVG of all players) x (your win rate)

AVG of all players ? How is that relevant ?
I stand by my point of using the -2.70 as where everyone should eventually end up at - vs 1.0 as used now for "breaking even"
Each player can be scored independently from the other players - I could care less if someone bets 1/2 million or 5 million while I'm betting 3k over more spins.
Looking at my 'serious' account vs those "above me" in the ranking.....
12 people "above me" have bet less spins then I have - even 4th place for example (yes - 4th place, near the top out of ALL players) has only bet 448 spins (less than mine) and his winrate is .995 vs mine (in profit) at 1.012
The only difference is that his bet per spin average is $492.50 and mine is $38.83
So I bet less, win more - have a better win rate and have played more spins..... he's 4th and I'm 30th.
Surely you see that this is flawed lol.
Using each person's personal standing against the -2.70 expected edge is how it should be done - not compared to other players.
That's how a ranking works - You don't take the gold medal from a runner because "on average" the other runners didn't make it around the track as far oO.

I explained how this is done - it only takes one additional set of data for each player that keeps track of amount bet total / played spins and compare that to the fixed edge the casino/game has built it and then rank the players accordingly.
I'll just keep track of my own stats and ignore the ranking chart for now, it makes no sense.

The data from roulette xtreme is mostly irrelevant, like highs, lows drawdowns etc. Again what matters is amount of bets placed, amount of spins played, amount won and amount lost.

I posted the picture to show the % stat that I highlighted - that's how to compare all players equally compared to the house - not against one another - how much they bet in $ won't matter (and it doesn't matter in the real world) because the % will show who's making bets that are beating the house edge and not just from luck.

The intention is again players who have the highest win rate AND have played comparatively more than others are ranked at the top.

No, as I stated above - I had done (in 30th) better than the player in 4th. More spins, better profit, higher winrate - the only difference is that person threw a lot more money down per bet - which isn't relevant in a ranking.

But to outrank others they'll need to make larger bets.

We're playing equally against the "house" - not "who can outbet who by making bigger bets"
Don't we want to bet the least amount, while winning the most - over time - consistently ?
Or do we want to rank people who bet 3 million $ on the table and are ahead $60.00 at the top ? that's nonsense.
Anyway lol - I'll keep track of my own stats. I count bet spins, amount bet and my bankroll.
From there I can tell exactly how much ahead I am ahead against the house edge.
I can manually do that with other players individually - but whoever programmed the site has the means to figure that easy stat out and rank players accordingly.
Again, just my 2 cents.

I agree with Turbo. It's a simple and fair ranking system.

Thanks

[Turner]

Turbo, you are taking this far too seriously me thinks lol

and...how can you play your way...in the future, with no past spins when all these spins are from the past.

That I dont get 

[RouletteGhost]

Turbo, you are taking this far too seriously me thinks lol

and...how can you play your way...in the future, with no past spins when all these spins are from the past.

That I dont get 

I don't understand it

betting hot numbers as you sit down is betting past spins

depends how you look at it

could be after you sit, then you start but you are still going off past spins

I sit down, I ignore the history board. 3 hits twice, now i bet 3 because it hit twice, those 2 hits are past spins

[TurboGenius]

Turbo, you are taking this far too seriously me thinks lol
and...how can you play your way...in the future, with no past spins when all these spins are from the past.
That I dont get

It's a competition for each player against the house edge - serious stuff.
And they aren't past spins, I'm seeing them for the first time. I don't look at past spins as I'm playing either (since it's clear when I start and log in)

[PeaBea65]

I don't understand it

betting hot numbers as you sit down is betting past spins

depends how you look at it

could be after you sit, then you start but you are still going off past spins

I sit down, I ignore the history board. 3 hits twice, now i bet 3 because it hit twice, those 2 hits are past spins

[Steve]

AVG of all players ? How is that relevant ?

It's relevant because you need to compare the RELEVANCY of amount of spins and bets made by each players. Otherwise short term lucky players will rank higher.

I stand by my point of using the -2.70 as where everyone should eventually end up at - vs 1.0 as used now for "breaking even"

This is almost exactly the same as the current "win ratio". It doesnt matter if we express wins vs loss as a percentage or value above or below 1. The difference is either -2.7% or 0.972972. You are asking to express the same win ratio in a different way.

Each player can be scored independently from the other players

Yes they can be. But the point of the leaderboard is a comparison.

Looking at my 'serious' account vs those "above me" in the ranking.....
12 people "above me" have bet less spins then I have - even 4th place for example (yes - 4th place, near the top out of ALL players) has only bet 448 spins (less than mine) and his winrate is .995 vs mine (in profit) at 1.012
The only difference is that his bet per spin average is $492.50 and mine is $38.83
So I bet less, win more - have a better win rate and have played more spins..... he's 4th and I'm 30th.

Yes but he has wagered more, so his results are more relevant. You are assuming his unit size is greater. But the truth may be that he's betting on more numbers. You might be betting $100 on 3 numbers per spin, and he may be betting $100 on 20 numbers per spin.

But I understand your point that if you were both flat betting on 3 numbers, and he bet $100 per number and you only bet $5 per number, then he would have an advantage in the ranking. From that perspective, it is a weakness in the algorithm. But as in real life, big bettors will usually win more if their system is successful. So if you are interested in ranking higher in the leaderboard, then just bet bigger. Nothing is stopping you from doing that. If betting on more or fewer numbers is an advantage, it will be reflected in your win rate and leaderboard rank.

Using each person's personal standing against the -2.70 expected edge is how it should be done - not compared to other players.

Again we already show the win rate, and it's really no different to the percentage you propose. Each player has their individual stats, but we must compare players to each other for leaderboard rankings.

That's how a ranking works - You don't take the gold medal from a runner because "on average" the other runners didn't make it around the track as far oO.

Any race involves comparison between other competitors. The only factor in a running race is time. But in the roulette game we need to consider other variables like amount of spins played, amount bet and win rate.

I explained how this is done - it only takes one additional set of data for each player that keeps track of amount bet total / played spins and compare that to the fixed edge the casino/game has built it and then rank the players accordingly.

But how does this solve the problem of lucky short term winners having a higher percentage, who then rank higher?

how much they bet in $ won't matter (and it doesn't matter in the real world) because the % will show who's making bets that are beating the house edge and not just from luck.

Unless we have each player set the value of "1 unit", we need to assume 1 unit = $1. It is the same for all players so nobody has an advantage or disadvantage.

I had done (in 30th) better than the player in 4th. More spins, better profit, higher winrate - the only difference is that person threw a lot more money down per bet - which isn't relevant in a ranking.

Yes but you dont know if they bet $1000 on 3 outside bets, or $3000 worth of inside bets covering lots of numbers.


My understanding of your points and summary response:

1. One player may make $1 flat bets, and another makes $100 flat bets. They can both have the same win rate, but the player with higher bets will rank higher on the leaderboard.

RESPONSE> Then just bet higher if you want to rank higher.

2. The more a player plays and bets, the higher they will rank even if their system loses.

RESPONSE> Yes but if their system loses, the more they play and the more they bet, their win rate will drop to expectation. So if someone plays the same amount of spins and wagers the same amount, but has a better win rate, they will naturally rank higher.

3. A player could just use a bot to automatically bet lots of money over lots of spins, and they'll get a 0.97 win rate and probably rank first.

RESPONSE> This would only work if other players played much fewer spins and had comparable win rates. This player would have no control over other people's amount of spins played, so they could have lots of spins played and the average wont change much. All it would take to beat such a player on the leaderboard would be a slightly higher win rate, and a reasonable amount of spins/bets.

Ultimately I do understand your point, but as far as I can see what you propose is no different to ranking based on win rate alone. That's how it initially was, but then short term players were ranking well above everyone else, then just stopped playing. And the harder players tried to catch them, the further they got from achieving top rank. If you offer a better solution and provide the mathematical equation, of course we'll code it.

Yes the win rate is the most important factor. But it doesnt tell the full story. The ranking algorithm MUST account for amount of spins played and amount wagered.

[Steve]

A simple enhancement could be:

Your rank score = A x B x C
                         = (your amount of spins played / AVG of all players) x (your amount bet / AVG of all players) x (2 x (your win rate))

Where A is never greater than 1, and B is never greater than 1.

The addition is in red. This makes the win rate more relevant than before. The effect will be more players who have good win rates but fewer spins played will rank at the top.

The ultimate solution would be to look at every individual bet, but would still need a comparison with other players, or short term lucky players will rank higher all the time.