Jump to content
** March Poker League Result : =1st Bridscott, =1st Like2Fish, 3rd avongirl **
** Cheltenham Tipster Competition Result : 1st Old codger, 2nd sirspread, 3rd Bathtime For Rupert **

Anytime Goalscorer Experiment


Recommended Posts

After 87 starting selections the data is starting to look a bit more realistic and reliable, with 30 scorers getting 36 goals.

Playing at the "true" odds implied by the spreads would have made for a return of plus or minus 8.19 points (with layers showing the profit).

Laying at the target lay odds would have made a profit of 14.93 points, thought I've only managed to get matched on 8 of the 87 players. I'm around £42.89 up from trying to lay each selection to £10 (probably slightly more as I sometimes have been matched at less than the target odds). So at least this hasn't been an entirely "paper trade" exercise.

Backing at the target back odds would have cost you just 0.31 points, suggesting perhaps that the back odds are actually closer to being the true odds!

Selling would have shown a profit of 308 points, buying a loss of 459 points. An average profit of 3.54 points per bet from selling should probably be achievable long term, even from selling at these lower prices.

I'm coming to the conclusion that I've identified a profitable laying strategy that may not be viable in terms of time and effort getting matched, but it may be possible to derive a less profitable but viable laying method.

I also think you're more likely to be matched at viable back odds on BF than lay odds. Anytime goalscorers is my most profitable fixed odds market so far this year, with just under half of the profit coming from the exchange.

Finally, it supports the view that player goal minutes is a viable market to be a seller in, even at lower prices. Outside of rugby union it's clearly my most profitable market as a seller.

So, all useful grist to the mill so far that supports certain assumptions but may help me refine my approach to laying, backing and selling.

Link to comment
Share on other sites

On 8/30/2021 at 1:20 PM, harry_rag said:

After 87 starting selections the data is starting to look a bit more realistic and reliable, with 30 scorers getting 36 goals.

After 104 (40 players scoring 48 goals) I'm starting to feel the data is less reliable again due to it having swung the other way! :eyes

Clearly starting with 17 players who all failed to score was unusual; my suspicion now is that we've got a higher than average number of goals being scored. Obviously it will help to get a bigger sample so I'll press on to at least 200.

What the data currently suggests is that the "true" price is fairly accurate whereas my experience suggests, in the long run, it will be too low. Backing at the true price would have pretty much seen you break even (-0.36 points with an ROI of -0.35%). I expected it to be a slightly bigger negative than that.

Laying at the "lay" odds and backing at the "back" odds (both with a 10% edge on the "true" price) would have been profitable. 8.8 points from laying with an ROI of 6.17% and 10.8 points from backing with an ROI of 10.38%. I suspect it would be much easier to get matched on the back than the lay (having only been matched on 12 players myself).

I think, with a big enough sample, I can get a better handle on using the spreads to frame a "true" fixed odds price. I just don't expect the starting notion of that true price to be as accurate as it currently looks (based on nearly 500 real money sells, albeit at higher prices).

Link to comment
Share on other sites

8 minutes ago, Torque said:

Bigger sample needed @harry_rag - whether you're ahead or behind right now :ok

Very much so; any idea what sort of sample size becomes trustworthy (dealing with roughly even money to 2/1 shots, average odds say 2.6)?

My guesses as to what the results might look like are backed up by many 100s of real money and paper traded bets but at shorter average odds. I'm interested to see where this leads and think it will be useful for me.

I've long used the spreads as a metric to identify possible value fixed odds prices which is definitely worthwhile but, having cottoned onto the wisdom of selling more often than buying, I've realised that the spreads are generally pitched slightly too high so I need to recalibrate.

Link to comment
Share on other sites

Thanks. 10,000 bets would be a stretch (I'm not as young as I was) but then I wouldn't carry on too long with anything only showing a potential 2% edge. 

I'm tracking multiple things here (true, lay and back odds) and I'm hoping it will throw up something that would show closer to a 10% edge (as backing at the "back" odds currently is), Oh, and that thing being something that is achievable in practice would be nice as well! So quite a modest ambition all told! :lol

Link to comment
Share on other sites

15 hours ago, harry_rag said:

I think, with a big enough sample, I can get a better handle on using the spreads to frame a "true" fixed odds price.

 

12 hours ago, harry_rag said:

Thanks. 10,000 bets would be a stretch (I'm not as young as I was)

Run a monte carlo simulation in excel,

You could do an estimated 10000 random sims for a year over a random 10 year span (100000 trials) on a decent computer in about 10~20 mins

Edited by Valiant Thor
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...