Day 4 started with my profit having been depleted once again down to £6.75 and day 4 finished with a small loss after backing all the days favourites. In this post I am going to look at grouping the races into a spread of distances to see how each group of distances faired. I am going to categorize the races into the following
Sprints – All races under 300m. Middle Distance – 300-599m Stayer races 600m and up
Lets start with what can perhaps be classed as sprints, ie all races under 300m On day 4 there were a total of 32 sprint races run one of which (Oxford 12:01) where no bet was struck due to the odds of the joint favourites being identical at the 20 second point. So from 31 sprints 13 favourites won producing a profit of £3.31 Middle Distance races produced an opposite result from sprints as the loss was £3.87 from 110 races however of these there were 19 bets struck which failed to be matched for whatever reason which has exposed a flaw in the software which needs to be addressed. There was just one stayer race which was at Monmore at 16:49 and this lost to a detriment of 70p stake money.
We now have a somewhat clearer picture of which races can potentially produce good results but this is from just one day If we include the previous 3 days into our figures we find that a loss of £2.42 is produced from sprints £4.21 profit from Middle distance races but this includes the non matching of those 19 races on day 4, and a loss of 83p from stayers.
I think that this shows the importance of collecting a large amount of data before risking bigger stakes, but we do have something to build on now and I look forward to analysing day 5.
Going into day 3 I have an overall profit of £9.99 over the first two days of the trial. If you have missed these posts then they can be seen by cliccking the link below Day 1 Day 2
there are also links at the bottom of the post.
Day 3 saw a total of 156 bets matched at betfair with an average stake value of 57p of which 52 dogs that were favourite 20 seconds from post time on betfair exchange went on to win their race. However this 33.3% of winning bets were not sufficient enough to produce a profit on the day. The figure was a loss of £3.24 – just 2 more winners would have seen a profit on the day.
The best performing track for favourites today was Suffolk Downs with a profit of £4.51 producing 7 winning favourites, 3 of these favourites came from trap 5 – another statistic which could help us in the future if you decide to pit wits in the trap challenge. but this is a subject for subsequent postings on this subject.
So after 3 days of the trial the 3 best performing tracks are Doncaster (£5.87) Romford (£5.24) Clonmel (£4.10) The 3 worst performing tracks are Towcester (-£3.46) Yarmouth (-£4.43) Sunderland (-£4.52) The full list is shown below
My total profit over the 3 days has now dwindled to £6.75 from £13.41 on day 1 (I wonder what the P/L would have been on the 1st of march if I had got my ducks in a row sooner. Having said that, making sure the software puts the correct bets in at the correct time is paramount especially if you are going to just leave it running throughout the day!
I am starting to build up some sort of data base now so I will begin to list best performing trap, distance and race class as we go along.
Best performing Trap (No of times trap Won over 3 days – not P/L) Trap 6 (won 82 times out of 467 races over 3 days) Best performing distance (No of Times favourite won over distance ) 480m (Favourite won 30 times out of 70 races run over 480m over 3 days) Best Performing Race Class A7 – Favourite won 17 times out of 30 A7 races run over 3 days producing a profit of £10.52! There were 7 A7 races run over a distance of 480m of which the favourite won 4 times producing a profit of £2.65 in the first 3 days of the trial.
As you can see the further I get into the trial I can build a picture of which race class and distance would be beneficial to focus on.
If you are following this series of posts you will know that after day 1 the profit from backing every favourite stands at a total of £13.41 with the best performing track being Doncaster with a profit of £4.73. On 3rd March there were a total of 13 Meetings at tracks across the UK with the best performing track being Crayford with a profit of £5.12 This is in direct contrast to yesterdays loss of £3.85. If you remember from yesterday I touched on the race class figures and Open Races proved to be more profitable In this case Crayford was profitable in the Open Race category but If all open races throughout the day were backed by favourites there would have been a loss of £3.02. The most profitable race class was A7 which produced a profit of £5.71
As you can see from a screen shot of my Excel workbook I can also break down the distance within the race classes to provide a clear picture of which distance, track, and race class shows promise. Once March 31st data is in I will make the workbook available for all In the meantime keep reading the blog.
My total P/L for 3rd of March was a Loss of £3.42 making a total profit across the 2 days of £9.99
For a while now I have been pondering the possibility of backing the favourite at the dogs to see if there is any money to be made. Even though the rumours that greyhound racing is not the most virtuous sport you can make a wager on it is certainly the most abundant sport with over 140 races each day and that is just the UK. According to the Oxford Stadium website 35% of all favourites win! This however is across all tracks, distances and, classes of race. To this end I decided that the only true way of sorting the wheat from the chaff is to back every single favourite in every single race and use the power of excel to weed out losing statistics from the more profitable. So without ado I will post day by day findings on this blog
My criteria is simple I will back each race favourite 20 seconds before the post time to a liability of just £1.00 so in effect my average stake should be kept to about 30p per race. This means that should the favourite win then the Profit will be £1.00 and if loses the stake determined by the odds that are available at that moment in time on the betfair exchange. There have been instances where the odds of two of the dogs have been exactly the same at the 20 second point and no bet was struck. This is unfortunate but one of those things. After testing that the software worked as required on March 1st testing began in earnest on March 2nd and showed an overall profit of £13.41 with profits showing from all meetings except Crayford and Hove.
Click to Enlarge
Starting with Central Park, there were a total of 6 winning favourites giving a small profit of £2.66 The rest I will put into a table for easy reference.
Track
Winning Favs
Avg Stake
P/L
Crayford
1
45p
-£3.85
Doncaster
9 (2 Meetings)
69p
£4.73
Hove
6
£1.43 **
-£0.51
Monmore
10 (2 Meetings)
57p
£2.44
Newcastle
6
48p
£1.69
Oxford
8 (2 Meetings
63p
£1.26
Perry Barr
8
47p
£1.48
Romford
10 (2 Meetings)
53p
£3.03
Swindon
4
49p
£0.48
** Quite a few Odds On Favs Lost!!!
Crayford was quite obviously not a track to back the favourite on this day as just one favourite won, Hove had a few odds on favourites running hence the average stake of £1.43 with the shortest priced favourite of the day losing costing a massive £3.12 All in all not a bad day for the first day. There were a total of 160 races of which 64 favourites won giving a 40% win rate.
Lets go on to look at the best performing meeting of the day which was Doncaster. Producing an overall profit of £4.73 with 9 winning favourites at the 2 meetings. Of the 8 winning favourites 4 were in OR class races or Open Class and scored a 100% strike rate as did the OR3 class which accounted for 2 more winning favs. The sheer volume of information I have collected just over the last 20 days is by no means enough and in subsequent posts I will delve a little deeper into which tracks are more profitable than others using this system and which Races should be avoided on all counts after looking at the stats.
Just as a Taster the best performing track using this system between 2 March and 22 march is Nottingham with a profit of £15.89 The worst Track for winning Favourites is Sunderland with a loss of £22.03
My total profit for this period backing every race is £38.28
Betfair Historical Data has been, for some time now, openly available for free but it seems that no one is prepared to share just how you open the damned stuff!!! There are plenty of sites out there willing to share their python knowledge but stop short of a full explanation – very frustrating. At this point I have to give credit to “Trading The Market” guys who have finally given me the Eureka Moment! If you follow the steps below you will soon have all the research information to back test your systems. To begin with you do need to have a betfair account and to access which data you need you will need to be logged in and at the following page https://historicdata.betfair.com/#/home
For the purposes of this post I will be using the basic plan which costs nothing and has a data frequency of 1 minute. If you need more frequent data logs then you will have to purchase your data by the month but there are free plans available for limited months mainly in 2020. In this post I will be analysing the data from the North London Derby – Arsenal v Tottenham on the 1st October 2022 The easy way to handle this data is to “purchase” a month at a time so I will “Buy” the Soccer data from OCT 2022 To do this simply select the plan, sport, and month you need in the drop down boxes and click “Add to Purchases” and then the “purchase” as in the screen shot below
You will then be asked to confirm your purchase – Check that you haven’t included the wrong plan by mistake as the fee will be taken straight from your betfair balance. If everything is ok click confirm and you will see the following screen.
As you can see from the screen shot there are over 250000 markets! for our research we only need 3 or 4 of these so we need to find them. I want to know how the under/over Markets behaved when each goal went in during the match. To do this click on the “My Data” Tab as shown below and you will see your most recent purchases as well as plans you have bought before.
We now need to populate the down load fields with the information that we need so if we look at the match stats we can see that there were 4 goals scored in total.
I want to see the following market odds O/U 1.5 O/U 2.5 O/U 3.5 O/U 4.5 O/U 5.5 O/U 6.5 and the Match. To do this we need to populate the download files panel as follows:
In the Plan Box (which is at the top but hidden in this screen shot) enter Basic In the Sport Box enter Soccer In the From and Upto Boxes enter 1 Oct 2022 Leave the event ID Blank In the Event Name box enter Arsenal In the Market Type box select the markets you need and in this case its Match_Odds Over_Under_25 Over_Under_35 Over_Under_45 Over_Under_55 Over_Under_65 In the Country Box select GB (There is no UK) and in the File Type Box select M Click the Add to Downloads button You now should have 12 files ready to down load. Click the Yellow Download button
You will find this file in your downloads folder as a Data “tar” file
Double click the file to un pack it in your unzipping program
Keep double clicking until you get to the actual data files
You should now see your 6 “bz2” files To make them easy to find Create a new folder on your desktop and name it ARS v TOT
You can now drag and drop the files into this folder from your unpacking software
We now need to convert these files into a readable format and Betfair provide one such program easily available from their website.
The first file we want to see is the Match Odds and this will usually be the largest file as most volume would have passed through that market. In this case we are looking at File 1.203213491
These are not very user friendly file names so I am going to rename them. to do this we simply select a file from the list using the betfair Historical Data Processor as below and note the file number and which market it contains. In this case File 1.203213498 is the O/U 6.5 goals Market. We dont want to do anything else yet. We are just identifying the files for our convenience
We can then repeat the process with all the files noting the file numbers to the markets and then rename all the files as below
We are now going to convert these files into meaningful excel ranges where we can look at them in more detail. To do this we again select a file from the folder using the Historical Data Processor and click the download settings button as shown in the screenshot below
This will open the settings dialog box as shown below
We now need to populate our spreadsheet with any relevant information we will need by clicking the drop down box arrow and selecting the fields by ticking the relevant tick boxes (dont worry if you put too much info into the spreadsheet as we can delete it later)
Once you have selected what you need close the drop down box by clicking the “up” arrow and because we are just interested in the in play data slide the “preplay” button to off as shown below and click save
We are now ready to download the file as a CSV file into Excel we do this by simply clicking the “Download CSV” button as shown and let the wonders of better brains than our own do its stuff.
This will put a zipped file into your downloads folder with the original file number
If we double click this file it will unpack and be available as a csv file which we can open with excel.
It may be a good idea to move this file to a new folder on your desktop and rename it as I have done here
If we open this in excel this is what we get.
We can now start to manipulate the data to best suit our needs. We can start by removing some of the columns we dont need. I have removed the following columns to be left with the following spreadsheet. Market ID Selection ID
The next task is to make the published time more reader friendly. as you can see this is total jibberish, it means nothing. If you select cell A2, in the address window you will see that the full date and time is shown
So that column A shows the time in hrs and minutes we need to select hh:mm from the custom cell formats as shown below.
When we click ok then cell A2 will show the actual time of the published data for that row
To convert all the row simply select Column A and repeat
When you click ok your spreadsheet should look like this
you will notice that the times are repeated 3 times this is because there are 3 separate outcomes within this betfair market – Arsenal to win The Draw and Tottenham to win. If we now draw our attention to Column F this is the last traded price column and we could do with tidying up this so that all the figures are shown to 2 decimal places. To do this simply select Column F and using the cell number format dropdown box, select “Number” as shown below.
Your spreadsheet should now look something like this
A bit more tidying up by changing the column headers and this is what you should be looking at.
This is your basic data spreadsheet for the entire match for the Match Odds Market You can now save it into a folder and repeat the process with the other 5 markets
In my next post I will show you how to manipulate the data so that we can see what happened in each market as each goal went in
Please like and subscribe to my blog so that I can get a feel for what content can be most useful to you.
With Wimbledon just around the corner I thought it might be interesting to explore the possibility of dutching the set betting market for the WTA as there is only 4 possible outcomes and using betangel’s dutching tool I can place orders into the market to ensure a 20% or better return on a £10 total stake. The french Open is currently at the 4th round stage in the WTA competition and there are just 4 matches today (Sunday 4th June 2023) The risk to reward on this is quite high depending on the odds offered. anything from a 50% risk up to a mighty 80% risk. In this exercise I will be placing the margin stake onto the favourites 2-0 score Match 1 Pavlyuchenkova v Mertens This is an evenly matched match where both players are at around even money to win the match. The screen shot below shows the bets placed
As you can see I have put the margin stake onto Mertens 2-0 scoreline at odds of 11 This is a massive risk to reward scenario as I am risking nearly 89% of my £10.00 stake for a 25% return. One of the factors that has helped me make this decision is that I have downloaded all the results from the previous 3 WTA French Opens into a spreadsheet and filtered the results into following criteria. 1. Winners price at the start of the match is between 1.90 and 2.10 2. 2-0 set result 3. 4th round matches only Having filtered these factors the resulting percentage of matches that ended 2-0 are just 3 from a total of 24 4th round matches – just 12.5% If we drill this down even further to include ALL matches that ended 2-0 where the winners odds were between 1.9 and 2.1 we find that from 381 WTA French Open matches that have been played in the previous 3 years there was just 20 that is just over 5% What we also have to remember is that we are only interested in the odds selected for movement only need to move so far in order to be matched. In close matches such as this a first set score of 5-5 could be enough to convince the market that the match could go to 3 sets.
As I said there are four 4th round matches today and the next one is between Muchiva and El Avanesyan Muchova is odds on favourite to win at odds of 1.27
In this dutching trade I am risking 53% of my stake for a 25% return While the risk to reward is much less than the previous match if we look at the spreadsheet and input the same figures in relation to this match ie where the Winning favourite was between 1.25 and 1.3 there was just 1 (one) match in the fourth round that ended 2-0 and just 21 out of all 381 French Open Matches from the previous 3 years.
The 3rd match of the day is between Svitolina and Kasatkina. Kasatkina is favourite at odds of 1.66 to win the match
The final match today is Stephens v Sabalenka where Sabalenka is 1.25 to win the match
With all dutch trades in place I will post the logs and final tally tomorrow My Maximum loss for these four trades should they all lose is £27.39 out of £40.00 My Maximum Profit is £10.17 This signifies a 37% reward to risk factor but considering the very basic research I explored I am hopeful for a 75% strike rate
When you embark upon trading any sport whether it is football, racing or tennis it is so easy to watch youtube videos of trading situations and think “this is easy” and jump straight into a market and lose your bank! While in depth research can be time consuming and at points seem repetitive some basic research should be undertaken if only to improve your chances of a better strike rate on your trades. The inescapable fact about trading is no matter how well you have researched a game, match or race you will not get a profit everytime, there will always be that odd market that will go against you. Everyone on Youtube bangs on about “Risk to Reward” and it is a very viable point and the first thing you should look at is exactly how far a market is likely to move against you should a worst case scenario happen – because like as not it will probably happen. For instance: If you are trading an Under 2.5 goals market in a football match you should try and establish what the probable odds would move to if an early goal was scored before your exit point was reached. How you do this is up to you but personally I use Soccer Mystic which is a great tool that comes with Bet Angel but that is a subject for another post but the essence is to know beforehand what two exit points you would employ should the trade go against you or if you are successful. What research material is available for tennis and in-play odds that will help you in this task. The main source that I use is Tennis Trader which is a software product that comes with BetAngel. Once configured to the match you are planning to trade on you can see at a glance at what odds the market would likely be trading at if, for instance, you were laying the server in a tennis match and he/she won the game. The screenshot below shows the GAME Matrix for the 1st Game in the match between Gracheva and Galfi in the 1st round of the WTA French Open
In this match Gracheva is odds on favourite to win the match at odds of 1.36. If we were to lay her as server in this first game for a £10 stake our liability would be just £3.60. We know by looking at this matrix and following a path of events (by this I mean the points won or lost) we can quite acuratly know that if she won the game then in all likelyhood the market would move against us by just 3 or 4 ticks – See below.
If, however, Galfi plays a stroke of genius and wins the first two points then the market would move in our favour by about 10 ticks. as shown below.
From this information we can now establish some exit points that we would feel comfortable with once we have worked out what our maximum loss is likely to be. To do this we can use a simple hedge calculator as found on GoalProfits.com To work out our likely loss on this trade we simple enter the figures as shown below:
This shows that if Gracheva wins her service game and we exit at the end of the game we are only likely to lose 23p. Now that we know our likely loss on the game we can choose our exit point should Galfi win a couple of points early on in the game by doing exactly the same using the figures on the bottom of the matrix.
If Galfi won the first 2 points in the game and exited the trade at Grachevas odds of 1.46 then our profit would be 67p
Once we have established our risk to reward we can have a look at the match itself and there are various websites available to us and a favourite of mine is tennisexplorer.com and what I am looking for is any previous Head to Head matches these two have played and what surfaces to determine any trends.
We can see that on clay (depicted by the brown square) in 2019 the sets were quite close until the final set. This could suggest that games swing to and fro in early stages of the match and some further investgation is warranted. Recent form can play a big part in the research and Flashscore.com can provide point by point results of previous matches. If we now look at the first set of Galfi’s recent matches and in particular the first few games we can get a feel for how the games progress.
In these 3 recent matches you can see in Galfis’ first service game she can be a bit unsettled with points bouncing between the players and while this is going on the market is moving suggesting a trade possibility. We can now look at the same thing about her opponent
In Gracheva’s first service games it suggests that she is more aggressive winning her games with the least amount of serves in the early stages of the match. The distinct possiblity of a back to lay trade on Grachevas first service game seems viable. This is just some basic research into one match. Lets have a look what happened.
Gracheva was serving first and won her service game with ease – just 4 points played. Below is the log of this game which was captured by Betangel automation and it shows the odds every 30 seconds throughout the match and how they bounce around. This particular screenshot shows the first 3 minutes of the match and as you can see Gracheva’s odds started at 1.36 and hit a low of 1.23 – a drop of 13 ticks.
I am laughing out loud now 🤣 because if you look at the begining of the post I showed you the likely odds would drop by just 3 or 4 ticks if the first game ended like this and I think that I didn’t calibrate tennis trader before I grabbed the screen – oops I stand by the suggested trade though as the basic research showed Gracheva to be an aggressive first game player and we could still have determined our likley exit point had the trade gone against us.
As a follow on to my third post in this series I thought it would be interesting to see just how much profit or loss I would make by dutching all the WTA singles matches from day 2 of the tournament using just the following criteria. Favourite starting odds of 1.20 or less £10.00 total stake Favourite to be backed at odds that would achieve a 10% profit margin
Dutching a Tennis match may seem a little bizarre as there are only 2 players but with the volatility of the market during in-play it is possible to trade in this way by backing the underdog at the outset at the same time as placing a back bet order into the market at odds higher than the current odds with the view that it will get matched at some point during the match. Some pre match research is essential when attempting this sort of trade so that you know roughly at what point your order is likely to be matched. This is done primarily with Bet Angels “Tennis Trader” tool. If you have read my previous posts on trading tennis you will know that this can be an invaluable way of predicting a feasible exit point for your trade and giving an insight to your potential risk to reward. It is a fact that in Tennis the WTA is the most prolific for having breaks of service, even with the most accomplished players and the fact that that players who are vastly overwhelmed by their opponents experience can overcome some of that prowess by playing in “overdrive” in the first set. This can mean that players who start the match at odds in the region of 1.05-1.2 can see their odds of winning the match increase to a point where this strategy can yield a 10% return on your stake. To illustrate this I have put a screenshot of the log I recorded yesterday of the match between Zidansek and Zheng. Zheng is seeded 19 and Zidansek is a qualifier.
As you can see Zheng started the match at odds of 1.16 but 10 minutes into the fist game her odds had risen to 1.34 before dropping back down to 1.18 and then back up to 1.3 and then back to 1.13. If we look at the actual game scores we can see that Zheng won the first two games breaking Zidansek’s serve in the second, but in the 3rd game Zidansek broke back to win the 3rd game and this is where these odds of 1.3 + would have occurred. This doesn’t mean that Zidansek was on to win the entire match it was just the market reacting to the in-play state of the match. Below is a screenshot of the first few games of the first set and as you can see Zidansek gave a fair account of herself.
So when we look at matches such as these “uncompetitive” matches as far as the starting odds would have it, there are some distinct possibilities that occur during the match and in particular in the first set that we can take advantage of. This is the essence of the trade I am going to outline in this post. If you have read my previous posts on tennis trading we are looking for the most reward against the least risk taken and to that end I am going to show you one match where there are two possible ways of making the same trade. For this I am going to turn to Bet Angel once again because it is the easiest way to place this trade into the market but this can easily be done manually as well by using a simple dutching calculator available for free on the internet. I will show this method later in the post. But first to Bet Angel. The screenshot below shows the Bet Angel dutching screen and the match loaded is between Vondrousova and Parks. Vondrousova is odds on to win with odds available to back at 1.10 Parks, the rank outsider is 10.00. We know that Vondrousova will 90 times out of 100 win this match, the odds tell us this so we would expect Parks odds to drift out to 1000 by the end of the match and we would be able to get any odds between 10.00 and 1000 matched at somepoint in the match so logic tells us that we should back Vondrousova at the start of the match and put an order into the market on Parks that is pretty certain to be matched at some point.
The Bet Angel Dutching tool enables us to select a margin of profit and total stakes If you look at the screen shot above the software is set to back Vondrousova for a stake of £9.99 and place an order into the market for a penny at odds of 1000 this, if it were possible, would net us 97p at the end of the match if Parks lost. The problem we have in this trade is that we would be risking £9.99 for a reward of 97p If we now switch this around so that we are backing Parks at the onset of the match and placing an order to be matched in play on Vondrousova the risk to reward has been turned upside down
This trade would now put an order into the market on Vondrousova at odds of 1.25 for £8.89 and place a bet that would be matched straight away on Parks for £1.11 at odds of 10.00. We now are risking just £1.11 for a reward of £1.09 – Even Money. So what are the chances of Vondrousovas back bet being matched in play? If we look at Tennis Trader (Screenshot below) We can see that it would only take a break of serve in Parks’ favour and for her to hold her serve until the 6th game for this bet to be matched.
It would not be advisable to blanket dutch all matches that have a favourite starting the match at odds of say 1.20 or less in this way as you would need to research the players to get some sort of feel that is the underdog capable of breaking the serve if he/she goes all out in the 1st set. But for the purposes of this post I will do just that on all of the WTA 1st round matches on day 2 in the French open today that have a favourite that starts at odds of 1.2 or less and post the results tomorrow.
Now I said earlier that you can perform this trade without the aid of Bet Angel by using a simple Dutching Calculator. For this I will use OddsMonkeys very capable dutching calculator which is available to use for free (I will post a link below) In the sceenshot below I have populated the fields with all the relevant information Stake £10 Parks Odds 10.00 Vondrousovas Odds 1.10 Commision 2% for both bets I have set the stake to be split over all selections and for the individual stakes to be rounded to a penny.
As you can see we need to adjust the odds for Vondrousova to make the profit show our 10% target for this trade which is about a quid.
1.20 is not enough as it only produces a 66p profit over all
1.3 is too much as it gives a profit of £1.43
Bingo! we have our desired odds at which we will place our order on Vondrousova and what stakes to use AFTER we have placed our bet on Parks. There is just one thing to remember and it is vital. Once you have placed your Vondrousova order into the market you must set the bet to “KEEP” otherwise when the market goes in play your order would be cancelled and you would in effect have a quid on Parks to win only. One more thing to remember is that if you are doing this manually on betfair your minimum stake is £1.00 so you may need to adjust your total stake to bring that to that amount.
Betornot 