The BTC Football Trading Thread
-
@matt-wood said in The BTC Football Trading Thread:
@darri I see, thanks again, looks like you have shown me that I have not understood something fairly fundamental from the start
Only taken me 7 months and you pointing it out for me to catch on 
I dont mind tracking the data. Should I be picking out one key line from a filter to track (ie the main driver for the filter results) or should I just track all of them? So taking the top filter that is similar to Lees, that would be the first 5 lines, so add in 5 columns for each line and start plopping the data in?
Thanks for this Darri I think you have just switched on a light bulb for me
@Lee-woodman I think you were trying to explain something similar to me on more than one occasion in the past. Thanks for taking the time to do that, just seems I am rather slow on the uptake
proper thicko at times If I were you I would add in a column for every single piece of data on your filter. I just export across and build my sheet around the CSV file. Track it exactly how it comes out. The only thing I remove is the ratio columns.
-
@matt-wood yes all, itll be a ball ache honest but itll also be very rewarding when you get to good data sizes and you can properly find the best trends and manipulate the data to produce a cracking filter for you to then trade. Once you get to that stage you then have a solid base of a filter to do as you please, be it set and forget or trade it inplay because the base filter is proven to work.
-
@darri I see, thanks again, looks like you have shown me that I have not understood something fairly fundamental from the start
Only taken me 7 months and you pointing it out for me to catch on I dont mind tracking the data. Should I be picking out one key line from a filter to track (ie the main driver for the filter results) or should I just track all of them? So taking the top filter that is similar to Lees, that would be the first 5 lines, so add in 5 columns for each line and start plopping the data in?
Thanks for this Darri I think you have just switched on a light bulb for me
@Lee-woodman I think you were trying to explain something similar to me on more than one occasion in the past. Thanks for taking the time to do that, just seems I am rather slow on the uptake
proper thicko at times -
@keith-anderson said in The BTC Football Trading Thread:
LTD / LCS in the Swansea v Brenford match. Currently 0-0 at HT
LCS in the Middlesbrough v Rotherham match. Currently 0-1 at HT
Back tomorrow. I promised my daughter I'd watch a film with her tonight!
Happy days! A bit of recovery from the Chelsea match
-
@matt-wood thats the way i do it if you look at my sheets, i actually dont track leagues probably should but i have a much smaller league pool than most because i target the highest liquidity leagues. But literally the only thing you need to add to that sheet you have already to make it easier to filter in future would be inputting what the stats are for each game.
So the top filter is very similar to lees and see you have included to adaptation of Home scored H2. All i would do here is simply track on your sheet next to each game is what stats these games are showing. Example, nt a filter just picked a random stat for the example:
All id simply do here is take maidenhead game for example. On the sheet id include everything you have already been doing but add a column with the stat input overall goals per match avg obviously make its shorter in title. But ll id do is input the stat 3.4. Do that for all the games and it will give a pretty easy to read trend for future when we come to looking to filter our results to see what i working or not. The layout of your sheet is fine just would highly recommend you add in column for the stats/filters. Make as many as you like yes its a ball ache but the more data we collect the more we can manipulate a filter to make it better. Obviously the change im suggesting isnt helping your current batch of results but im hesitant to make rash changes without also knowing key stats.
-
@thomas-larsen GOAL !! 0-1 and greenen up
-
@darri Thanks Darri. I think if I am understanding you correctly then you are pointing to what seems to be a gap in my knowledge.
I am not sure I can tell you what on my filter is driving the games that come up. These are my filters, they are very heavily inspired by yours and Lees. I think initially I did them to see how it worked but perhaps I skipped over some learning at the start.
So what I should have been doing from the start is picking out the key line from these filters and tracking that along with the games to see how successful it is. Then after than I should be moving to where my sheets are now?
Apologies all, not sure how I make the pics smaller
-
LTD in Lask 2.70
-
@matt-wood just had a look at the sheet to see if i could help. There is a a pretty big thing missing tho. Thats what stats your using to find games from the stats software. So next to each game id look to add another column with your stats software key stats. Its what lee was talking about. You look at my sheet: match, key stat, score, goal times, w/l, odds. Now of course adding in more like what the other guys do would mean more angles but for me im solely focused on SHGs. Id keep it stupid simple. Is this games coming direct from a filter? If it is then collecting the data on what stat inputs your using will help filter out games. Right now with yours really its only leagues and HT scores/who scored etc that are the main things to breakdown/measure. Im building a new SHG filter to further improve my late goals strategy, id like to be increasing it by about 3-5% strike rate so im slowly building up data to see what stats/scorelines actually matter when looking for a 2nd half goal. Take a look at my sheets or my blog when i did a breakdown ill quote it on my blog after this and have a look at lees 2H filter page where he showed his breakdown, because since your looking at shg filter then we both have good experience and would be able to help you bud. Least your being honest and seeking help.
Example would be if your using away concede by FT => 80% then mark down on the sheet in a column headed with the away conceded and mark in what stat that comes up as in stats software. This will allow you to further filter your picks and help build a filter. The data your collecting is actually data id class as fine tuning and last stage filtering ie leagues and scores. I think early days is all about testing the filter then that stuff. You know where i am if you need help bud. I know lee said there are other filter for SHG out there already but it makes it easy to understand when its your own work so put in these changes and keep going. I would run down and look at scores and leagues etc but actually you might be deleting leagues without first having an optimised filter setting if you get what i mean.
-
LTD / LCS in the Swansea v Brenford match. Currently 0-0 at HT
LCS in the Middlesbrough v Rotherham match. Currently 0-1 at HT
Back tomorrow. I promised my daughter I'd watch a film with her tonight!
le jeu avant tout
Am Yisrael Chai
When you got skin in the game, you stay in the game
But you don’t get a win unless you play in the game... -
@keith-anderson said in The BTC Football Trading Thread:
LTD / LCS in the Chelsea v Wolves match. Currently 0-0 at HT. I’m still in the lay Under 1.5 goals market too.
Terrible
-
Evening guys, small LTD- LCS at Chelsea on 75. 1.73 and 2.24.
-
LTD / LCS in the Chelsea v Wolves match. Currently 0-0 at HT. I’m still in the lay Under 1.5 goals market too.
le jeu avant tout
Am Yisrael Chai
When you got skin in the game, you stay in the game
But you don’t get a win unless you play in the game... -
@matt-wood said in The BTC Football Trading Thread:
Thanks guys, I think I have added the spreadsheets to my signature so anyone can feel free to poke at it and pass comment
Looking forward to the backtesting facility being added so we can tinker at will without having to think.....s**t, back to the drawing board haha!
-
@matt-wood I would look at it from as many angles as you can, like Stuart said. Often it throws up other stuff. My SH filter was originally intended to be a 2 SH filter. I appreciate it’s not too dissimilar but there may be something there. Write down as many stats as you can think of and see what the % SRs are and go from there. You may find it’s not great for 70+ but has a good SR if a goal has been scored already, so you could trade that one of 2 ways. I’d look at removing poorer leagues and break down score lines.
Whenever I record a filter I always add extra criteria (which I set to zero). It adds no time to export the extra data but allows me to add in extra layers if necessary or switch one criteria for another. Almost every filter I do has between 15 and 20 criteria recorded
If you are after a SH goal filter there are a few knocking around, maybe just use one of them and see how it can work for you
-
-
@keith-anderson said in The BTC Football Trading Thread:
@matt-wood said in The BTC Football Trading Thread:
@stuart-capstick might have to spend some time looking at it from other angles then
Yes, you really should. Or get someone else to look at it. They may spot something. A lot of strategies are just recycled or upcycled work belonging to someone else.
Nailed it, Keith.
-
@matt-wood said in The BTC Football Trading Thread:
@stuart-capstick might have to spend some time looking at it from other angles then
Yes, you really should. Or get someone else to look at it. They may spot something. A lot of strategies are just recycled or upcycled work belonging to someone else.
-
@stuart-capstick said in The BTC Football Trading Thread:
@matt-wood said in The BTC Football Trading Thread:
Question for the more experienced hands on here:
What do you do if you reach a point of your collected data not meeting the required standards. For me I mean that my 2HG filters are maybe not quite good enough (combined with my lack of ability to correctly read a game) but I have over 500 games on each filter. What to do with that now?
Have a closer look at the data. Maybe no good for 70+goal, but may be profitable at U2.5 or O2.5 or FHG. Or laying U1.5 from 15 mins.
By sheer chance, a failing FHG filter turned into a very good U2.5 filter.You've good 500+games. That's probably enough to give you an idea if you are onto something. Then you can possibly collect other variables.
I guarantee somewhere in your dataset there'll be something that nudges you towards a profitable strategy.
To be honest with mine, it's finding plenty of goals where goals have already been scored early doors so it might just be a ccase that it will do something different to what I first expected.
Won't know unless I keep putting the work in.
-
@stuart-capstick might have to spend some time looking at it from other angles then
