Jump to content

Calculating Odds


Recommended Posts

Hey, I read a few articles I found posted on here about Value and determining what bets are value based on working out your own odds. I also read an article about ratings and using a point allocation system for creating your own. Does anyone know of a structured way I could convert my ratings into odds for each of the runners in a race?

Link to comment
Share on other sites

Re: Calculating Odds Let's say you rate 5 horses 1) 52 points 2) 37 points 3) 16 points 4) 7 points 5) 3 points Add the total amounts of points up - in this case 115 points divide horse points by total points. 1) 52/115 = 45.22% 2) 37/115 = 32.17% 3) 16/115 = 13.91% 4) 7/115 = 6.09% 5) 3/115 = 2.61% Hope that helps :ok

Link to comment
Share on other sites

Re: Calculating Odds To go from probability to a price the formula is ((1/prob)*100)-1 So for the above example :- 1) (1/45.22)*100)-1 = 1.21 2) (1/32.17)*100)-1 = 2.11 3) (1/13.91)*100)-1 = 6.19 4) (1/6.09)*100)-1 = 15.42 5) (1/2.61)*100)-1 = 37.31 :ok

Link to comment
Share on other sites

Re: Calculating Odds Thanks Kanga, thats a great help, basic mathematics I suppose. Thats looks pretty cool if im understanding it right. Lets see. In this example I have calculated ratings using a point based system for a 5 horse race. Horse 4 was rated 7pts which converts to a 6.09% chance of winning. If I wanted to bet on this horse but wanted to ensure value, if I was to look on betfair I should be looking to match 15.42 (decimal) or at someone like Hills 14/1 (rounded down for simplicity) or above.

Link to comment
Share on other sites

Re: Calculating Odds I think this is where I struggle with the Value based approach to betting. I guess I could use this approach for pricing up races and if I find that the three top horses in a race have similar ratings thus chance of winning according to my system then when looking at the markets I would be wise to bet on the one offering the better value i.e the one being offered by the bookie at a far greater price than the other two. As long as I have confidence in my system then this should get me value. I find the term value quite subjective, if someone offers 1000 to one on a horse which I think should be 150/1 is that still value? People use the term value in a way which is taken to mean by getting value over the long term you should win more than you lose but im pretty sure if I kept backing 150/1 shots at 1000/1 I would still lose out over time.

Link to comment
Share on other sites

Re: Calculating Odds

In this example I have calculated ratings using a point based system for a 5 horse race. Horse 4 was rated 7pts which converts to a 6.09% chance of winning. If I wanted to bet on this horse but wanted to ensure value, if I was to look on betfair I should be looking to match 15.42 (decimal) or at someone like Hills 14/1 (rounded down for simplicity) or above.
In the example of horse (4) the "fair odds" would be 15.42/1 at bookmakers odds or 16.42 on the exchanges (real life prices would be 16/1 but you know where I'm coming from) - this would mean that if your pricing up a race was accurate over time you would break even at that price. When you look for "value" you are looking to beat the "fair price" - you can go either way with this - if this horse was on offer for 8/1 you should be looking to LAY this price - if the price was 25/1 you should look to BACK at this price. How much "value" you are looking for is a personal thing.
Link to comment
Share on other sites

Re: Calculating Odds

I find the term value quite subjective, if someone offers 1000 to one on a horse which I think should be 150/1 is that still value? People use the term value in a way which is taken to mean by getting value over the long term you should win more than you lose but im pretty sure if I kept backing 150/1 shots at 1000/1 I would still lose out over time.
As long as your rating are accurate and you were offered 1000/1 on a 150/1 shot then yes that is tremendous value and over time you would make a very nice profit (in a perfect world) on each 151st attempt you would get 1000 back. In reality you could get 2 results like that together then have to wait 2,000 more attempts before another one and a rare few would have that amount of patience, dedication and belief. Read this link it changed my betting totally. http://www.cdsystems.uk.com/value.htm :ok
Link to comment
Share on other sites

Re: Calculating Odds Ok had a little experiment tonight, chose the 21:00 at Hamilton only 4 runners and gave me time to prepare. First thing I got the current best price odds from attr. 11/10 Home Sweet Home 5/4 Just Lille 6/1 News of the Day 50/1 Craig Y Nos I figured the Bookies think the horses chances of winning are as follows. 90% 80% 16% 2% I then set about forming my own ratings using this approach http://www.cdsystems.uk.com/rating.htm I found the ratings to be Home Sweet Home 35 Just Lille 55 News of the Day 55 Craig Y Nos 15 This meant I figured the % chances of winning 24 % 37 % 37% 2% Equating to odds required of 4.3/1 2.6/1 2.6/1 50/1 By now it was approaching off time. According to my calculations Home Sweet Home was vastly underpriced and I managed to Lay it at 1.99 on the exchange. I rated Just Lille and News of the Day equally but with Just Lille slightly under priced but News of the Day offering Massive value now at 7/1 I chose that as the horse to back. Whilst my calculations agreed with the bookmaker that Craig Y Nos only had a 2% chance of winning and should be priced 50/1 I also had a couple of quid with betfair who offered 130/1 exchange which also represented value. So heres how it panned out 1. Just Lille chased leader, ridden to lead over 1f out, driven out 2. Home Sweet Home chased leaders, ridden over 1f out, chased winner inside final furlong, one pace 3. News of the Day led, ridden and headed over 1f out, weakened inside final furlong 4. Craig Y Nos always behind I was right in that Just Lille was very capable of winning and that was the case, unfortunately the value was not there. I was right to Lay Home sweet home due to the chance of it winning in my eyes being underpriced. Although I fancied News of the Day to go close and it offering value it proved not to be the case. The same with the chance taken on Craig Y Nos. However the Lay covered the money staked on News of the Day so only the small stakes lost on the outsider. Not sure where to go from here really. Should I focus on a particular type of race i.e track type, runners, distance, age, class. Then try to improve the rating method? I want to neaten the system up so its fully structured. Any advice welcome.

Link to comment
Share on other sites

Re: Calculating Odds Well done on your calculations !! My advice would be to paper trial different rating sources for your calculations. If it gets too much just try one meeting a day. The hard work is in finding a reliable method but once you do you will profit.

Link to comment
Share on other sites

Re: Calculating Odds

First thing I got the current best price odds from attr. 11/10 Home Sweet Home 5/4 Just Lille 6/1 News of the Day 50/1 Craig Y Nos I figured the Bookies think the horses chances of winning are as follows. 90% 80% 16% 2%
If you add up the %'s they should equal 100% The formula for converting odds to a % is 1/(odds + 1) * 100 (1) 1/(1.1 + 1) * 100 = 47.62% (2) 1/(1.25 + 1) * 100 = 44.44% (3) 1/(6 + 1) * 100 = 14.26% (4) 1/(50 + 1) * 100 = 1.96% In this case the % is over 100% because they were the best prices available from all bookmakers - if you were to calculate the prices from 1 bookmaker it WILL be below 100% and usually around 92-94% mark so they make a profit on whatever outcome.
Link to comment
Share on other sites

Re: Calculating Odds

If you add up the %'s they should equal 100% The formula for converting odds to a % is 1/(odds + 1) * 100 (1) 1/(1.1 + 1) * 100 = 47.62% (2) 1/(1.25 + 1) * 100 = 44.44% (3) 1/(6 + 1) * 100 = 14.26% (4) 1/(50 + 1) * 100 = 1.96% In this case the % is over 100% because they were the best prices available from all bookmakers - if you were to calculate the prices from 1 bookmaker it WILL be below 100% and usually around 92-94% mark so they make a profit on whatever outcome.
Ah thanks for that I did wonder why my figures diodnt add up. I will try again on Saturday so I can spend some time selecting which race and looking at the form for a bigger field.
Link to comment
Share on other sites

Re: Calculating Odds Your formula Kanga looks a bit confusing. I prefur to say: First figure + second figure = resultant figure. Second figure divided by resultant figure = decimal Convert decimal to percentage. e.g. 11/4 = 11 + 4 = 15 4 divided by 15 = 26.7 Therefore 11/4 = 26.7% Anything you calculate to have a 27% chance should be backed at greater than 11/4. Must say I am very encouraged by members of this site. Used to use a rival one that was mainly system junkies. This is much better. Ginge

Link to comment
Share on other sites

Re: Calculating Odds I seem to be getting differences between the two methods, I must be doing one of them wrong? Trying to figure out which and why?

Runner Top Frac Bottom Frac Top Frac/Bottom Frac + 1 = Decimal % Chance=(1/(decimal+1)*100)
Kangas way
1 15 8 2.88 25.81
2 9 4 3.25 23.53
3 5 1 6.5 13.33
4 5 1 6 14.29
5 12 1 13 7.14
6 25 1 26 3.7
Total 87.8
Gingertipsters way Top Frac Bottom Frac Top Frac + Bottom Frac %Chance =(Bottom Frac/TopFrac+Bottom Frac)
1 15 8 23 0.65
2 9 4 13 0.29
3 5 1 6 0.14
4 5 1 6 0.14
5 12 1 13 0.07
6 25 1 26 0.04
Total 1.33
Link to comment
Share on other sites

Re: Calculating Odds Windsor 21:10 Azreme I rated 50 = 22% probabliity = odds required 3.55 Corlough Mountain I rated 45 = 20% Probablitiy = odds required 4 Valley Observer I rated 25 = 11% probablity = odds required 8.09 Prince of Charm I rated 60 = 26% probablity = odds required 2.88 Divine White I rated 39 = 17% probablity = odds required 4.88 Salisbury Plain I rated 10 = 4% probablity = odds required 24 I opted to back the Prince of charm @4.2 exchange to place, as this is where I saw some value. Acording to my calculations all the other horses were over priced. The result however did not make good reading.

1 Azreme (6) 7/4 Fav 7 10 - 0 S Drowne P Howling 61
tracked leading pair, ridden to challenge 2f out, led over 1f out, driven and in command final 100 yards
2 3/4 Valley Observer (FR) (5) 4/1 3 9 - 9es1 A Kirby W R Swinburn 65
tracked leader, led 3f out, strongly pressed and headed over 1f out, stayed challenging, held final 100 yards
3 3 Corlough Mountain (2) 9/4 2Fav 3 9 - 9 William Buick(5) N A Callaghan 70
held up behind leaders in 4th, went 3rd 3f out, effort on outer and wandered under pressure over 2f out, one pace and no impression from over 1f out
4 2 1/2 Salisbury Plain (7) 16/1 6 8 - 11 Nicol Polli(5) N I M Rossiter 49
took keen hold, held up in 5th, ridden and one pace final 2f
5 9 Divine White (1) 11/1 4 9 - 5 R Hughes P Bowen 52
led to 3f out, soon weakened
6 17 Prince of Charm (USA) (3) 7/1 3 10 - 0p G Baker R A Teal 68
always last, struggling final 3f, tailed off
Jools (4) Non Runner 53
I think I may alter my rating system to give more emphasis to the Ground the horse has been succesful on before compared to todays going and also the type of course as I heard it going is the single most important factor. I need to look at how/why I got Prince of Charm rated far too high????????
Link to comment
Share on other sites

Re: Calculating Odds FG - You always get a better 'feel' for a subject when you understand the mechanics behind it, but you could possibly save some time and reduce the likelyhood of making mistakes by setting up a spreadsheet to do the calculations automatically or by using a 'tool' such as this :- http://ozracetools.com/modules.php?name=Poll_Points Just enter your ratings for each horse in the poll points colunm and it will calculate your odds (decimal, ie 3/1 = 4). Hope this helps.

Link to comment
Share on other sites

Re: Calculating Odds This table is called the Table Of Odds And Chances. Once you have done the calculation once and made a note of it, thats it. You do not have to do it again. e.g. Once you know 9/4 = 30.8% you can look at a bookies board, see the price of 9/4 and immediately know it should only be backed if you believe it to have a better than 30.8% chance. It may be best though at least at first to restrict your betting to those a fair bit above your estimate. e.g. only back 9/4 shots if you think it has a 35% or more chance. There is no need to know the exact percentage, just call 8/1 11% instead of 11.11% for example. Once you know the table off by heart and can see bookies prices in percentages it makes finding value a lot easier. I have the table with a full explanation on my website. Do not worry it is totally free and I am not trying to sell you anything. I suppose I am allowed to tell you it. www.gingeracegoer.plus.com ginge

Link to comment
Share on other sites

Re: Calculating Odds Thanks gingertipster thats an excellent webpage, really useful info and am sure will be of use to many other new-comers to here. I now need to cencentrate on working out whhat percent chance the horses have of winning from studying the form. I like the kindo of system based approch I got from http://www.cdsystems.uk.com/rating.htm but im not convinced about the entire method as it is throwing up some oddities. I will paper trail my findings as I tweek the system into what I consider acceptable ratings.

Link to comment
Share on other sites

Re: Calculating Odds Hi FG, Just remember that if you give a selection a 25% chance of winning then don't expect to get it right every race, it's easy to forget that your selection will only come in proportional to the % you've calculated. Also they won't come in convenient 0001 0001 0001 bundles :loon. Paper trading is the way to go, and the current wild swings in going hasn't made things any easier lately. Try picking an A/W course let's say Wolves as IMHO this is nice consistent course to trial. :ok

Link to comment
Share on other sites

  • 3 months later...

Re: Calculating Odds

Converting ratings into odds: http://www.twonix.com/twonix/library/Ratings_Odds_1.PDF The only one i found. Any other ideas? Dices
Dices, did you trial this method for any ratings ? Ive just had a read of it and i can get to the same figures in the table, but there must be a better way than i have done it, can someone check how i have done it please, my Math stinks :$ :$ :$ Using the figures from his table in figure 3 i did it this way ? Horse A (22 + 18 + 12) = 52 52 * 6 ratings = 312 312 + 96 deadheats = 408 1200 - 408 = 792 792 / 3 = 264 1200 - (408 + 264) = 528 528 / 4 = 132 1200 - (408 + 264 + 132) = 396
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...