Runnerjp - My Journey!

[A Former User]

It's starting to look very good mate and stream lining everything is going to make your life easier. I have been doing a fair bit of listening/reading lately to a lot of pro punters about modelling and the thing I have taken from them is that simple models just do not work and you have to put together very complicated systems to make money.

I would be interested to get your and others thoughts on this as the models such as Ryan and Martins filter that get sent are not complicated and have been working well.

[A Former User]

So here we go....

I have created a script that now tracks Live bets and trades that are in place. The reason was I found myself checking Betfair app when bored ... shopping, adverts on ect and what I want to do is get way from opening the app all together at the moment - no side bets ... no temptation to double up ect. So as soon as the database picks up a tade or bet it displays in this panel. Great if I am wanted to watch a game and cant decide!

One thing I forgot to mention is all outlay per start so I will add these next to them.

Onto the week.....

This start uses £2 - could make higher as does not come about often atm due to bloody cup games but slow and steady. Nothing has matched this week so stable...

Teams lays is progressing well and uses £1 stakes this only activates second half and lays the leading team for £1 risk currently. Started off in the negatives and has shown an upward progression

My back are just floating and I have had a look back at my trail result and after abit of digging I am going to remove these. this is because the ruturns can be really poor and 1 loss wipes it out 2 quick. I was lucky in testing that it picked a couple of high odds games that boosted profits. I am going to stop this as I have something else I will mention at the end.

Last of all my golden egg!!
This is a little aggressive but still using £1 stakes the liability can be up and down ... lows of evens ... high of 8's! but it can take hits and I am happy with the risk/reward. This is the money maker but may be longer to scale up due to the liability.

£20 in 7 days is nothing to be sniffed at especially when its been blank for 2 days in the week due to cup games! Gets abit mental at weekends but fun to follow

I have added John's Trobs and thats now auto setup but the bloody BOT keeps picking up its Heavy -> I wont argue with it as I want this to be truly automated so I will let it run and report when it placed a trade.

So my current overall looks like this... I am happy with the graph so far!

Now I have something that replaces my Backs ... I will need to update fully later as wife wants to go out but....

I said I was swapping my live platform to my VM so its on it own with no interference ... as i did that i forgot i had a couple of starts running in test mode for a month - couple where crap ... a couple where like the lays and gaining abit of progress but one of them was killing it! I looked through the results and I was very happy with what i saw ... i activated it yesterday and smashing some profit from the off.... i will swap up the backs with it in my charts and update on it later.

Sorry about spelling mistakes but i am in a rush before wife spots me on here rather then getting the 3 year old ready to go out!!!!

[Martin]

@jarratt-perkins let us know how it goes! always good to see things tested by others

[A Former User]

moved it over to its own VM now as I finally got the firewall to unblock the data collection.

My week runs sat - fri as weekends are busy its nice to see where you stand for rest of the week! I will update my profit loss and write up tomorrow

[A Former User]

@martin-futter said in Runnerjp - My Journey!:

@jarratt-perkins love this

I have also added Johns TROBS as well so have everything I want not to track .. moving onto the bigger side now.... modeling!!!

[Martin]

@jarratt-perkins love this

[A Former User]

Managed to get it working in Live mode so I can take a gander or what it's picked

I think what I am trying to show hear is a line graph is probably the best way to go to have confidence in what you're doing. I think its better then knowing how much you have in the bank

Any questions about anything from ho I got to this point - setup - coding help ect please ask away

[A Former User]

Ok we are up and running!!!

Welcome to my admin portal (still in progress)

These are the 4 Start I am now using (I had 5 but found a lot of the trades I was doing where duplicated within it so removed it.

Quick Recap

Armageddon!! (Used to be Goals Goal Goals but took up 2 much room lol) - This looks at halftime and has specific set of values to be matched and then stat betting and trading between the draw and over 1.5 over alot of points and dependent on how the action continues to flow.

Teams Back and Lays - This looks at pre stats and also inplay stats at HT to see if a team should be backed or layed (these can be 0-0,1-0's or even 2-0's)

A Boring End - This one has come from my original idea of backing under in extra time when nothing was happening. I have extended this to 85 mins as it grabs more profits and has restricted to specific leagues!

HOW IT WORKS!!!
I thought I would explain how all this works incase anyone wanted to copy (note it's taken me a month of hard work to code!!!).

BF Bot places all the bets and trades ect auto - I do nothing here now. Once a file in the C drive gets updates (when a bet is matched or settled) it updates the file timestamp (Anyone intrested the zip file is here - C:/Users//AppData/Local/bfbotmanager.com/Bf Bot Manager V3/uk_bets_history.gz). I then run a script in the background that keeps an eye on the file and if the timestamp alters it kicks into action and unzips the file and writes it to a local sql database.

Then I can write sql querys to output profit loss ect depending on the start I am using.

Now this is the best part for me. I can stick it into graphs and have pin points to look at where i was and where I am now. I have removed the dates as tbh I dont really want to see anything more than my graphs going up! Along as they are going up then I know they are working.

Whats Next

Nags is next on the list and I will continue working on the dashboard. I have spun up anouther Vm to add BFB into Sim mode to start with the nags. I also want to start building upon the models I talked about before.

I will be now posting up these screenshots every Friday - Sunday when ever im free to show progression - Lessons learnt ect.

[A Former User]

Little delay on this - works mad and i only went and wiped aload of the data on my ruddy server so the bfbot manager has buggered up alittle. Might have to start from £0 again as data all a mess >

[A Former User]

@darri I think you make a great point "your trading style is different to mine". Its about finding what works for you and you can control.

As working in IT I prefer to code and put it all into automation whereas others may not like that approach and want to be hands-on ect .

The main aim for all ... make green

[A Former User]

@jarratt-perkins good luck bud, really looking forward to reading these updates as your trading style is different to mine itll be a good insight into how automation works and your layout of posts are really informative.

[A Former User]

OK so today is - I have gone back live day and starting from 0!

Generally, people don't put how much their bank is or how much points are ect but I am very transparent and want to grow my bank and show a large bank is not needed so - £100 bang on start

[A Former User]

@darri Hi Mate - Didn't spot the reply as your D logo is green like my J so though it was just me typing at quick glance lol.

So I have actually trimmed it RIGHT down to 6 starts from 15 A lot of them were picking the same games just applying slightly different methods so it was an easy decision to chop them down.

Funny you should say about UFT and profit ... only had them and DFT running for 2 weeks and I have already made the decision to cut them. I was running them at low stakes in live as test to see how they perform and not 1 start made a profit on a daily basis. I would go for longer but when you break them down today by day and only 1 or 2 days had green you know it's just not working. It's not the start itself as all the starts we use actually are basically the same just diff entry points or markets used. It is in fact the selection process that the issue.

What I must say is after messing with them all I have actually created an evolved start from using The Relay method and match selection process from DFT so create 2 bots that work off one another (backing and laying) and using one of the most important things I think we ALL need to be using and that's in play stats!

I think these 6 starts (it's actually 5 as one is just the opposite of the other (back/lay) but want to keep them sep to ensure they are performing to the best possible outcome.

Only thing I am doing is pulling result and checking what the pre-match stats were for the game and seeing if a can see any patterns along with price entry. I will go into more around the actually starts now and go from there.

[A Former User]

@jarratt-perkins awesome stuff here mate, i think your a brave man to be trading 15 strats, iv not heard any of the pros doing such a thing, but i think the way your going about it i think youll come out fine. I suppose that why you are an advocate for automation. Out of interest are your own strategies more profitable than say the UFT ones? I tend to find my own strats perform alot better than paid for ones because iv learned how to use them more.

[A Former User]

Looking to post updates every Friday now to keep it weekly but wanted to add my main update in prep for then.

I have done some tidying up and cutting and sticking together to give my self 15 football starts that are in play. The MIx of pre-match and in-play selections so wanted to break them down today ready for Friday.

UFT - I have gone with the following 3 which are all automated from selection to trading!

The Fireball
The Snowball
The Relay

DFT - Once again all automated from selection - to excel spreadsheet to BFbot Man.

H Bomb
Time Bomb
Armagedan
Core
Sticky
Fireball

And now my own.

OWL - This looks at inplay stats and pre match odds/stats and places a bet on o 2.5 and 1-1. It will asses if required to cover some 2-0 0-2 at ht!

Goal Goal Goal - This looks at halftime and has specific critered to be matched and then stat betting and trading between the draw and over 1.5 over alot of points and dependent on how the action continues to flow.

Boom Train - Once again a 2nd half start that trades overs an under depending on the action!

Aussy FH - Simples .. back fh goals!

A Boring End - This one has come from my original idea of backing under in extra time when nothing was happening. I have extended this to 85 mins as it grabs more profits and has restricted to specific leagues!

2nd half Back and Lays! - These are my busy ones and i have had to keep stakes low as gets CRAZY on a busy night at HT looks at how teams are performing and looks to either back or lay those teams.

Those are the starts and it looks like ALOT which it is but if a say Aussy only happens Fri Sat Sunday in the morning, the 0-0 have only had 3 matches in last 2 weeks and UFT and DFT are all dependent on match status, price then it's not a lot of volume for trades.

I might tighten some of these up even more by making sure the pre-match selection fits specific criteria and THEN look at in play even more but I don't want to tie down something that's currently working.

Like I say its now ALL automated which i will review again on friday so not a thing for me to do but do some validation after games to see if there is anything i am missing to improve further. Dashboards will be created for me to track them all soon

See you all Fri!

[A Former User]

So 7 days in and here is what it looks like! I still need to build my dashboard but not had the time.

My Aussy Bot is doing well with just over 2 points profit with 100% strike rate so far over 2 games. This take into account a few factors and if the price is right-backs overs0.5 fh at kick off and at 15mins if the price is still looking good.

0-0 85MIN BOTS

These look at the scores at 85min and if 0-0 and meet a set criteria will back the 0-0.

So I stupidly reset the prem league the other day so it shows £0 but its also showing profit. Again 100% strike rate so far.

These are my second half bots which have not started well but i know will pick up!

I do have some other bots that are doing well BUT are UFT ones that I know people will have paid for so not sure I should stickup or not so far now will leave off.

So I have another 100 BOTS running (no joke) that are looking at multiple factors and playing off one another (some post fake bets to allow me to know when goals are scored ect to build up an inplay picture such as favs scoring early and still leading at ht).
So these are split into 2 starts - 1 is when a team is leading at ht my bots back them depending on multiple factors hence a large amount of bots. For example, it has just backed the Perth game at 2-0 for pert to win!

1.3 is cracking price for a 2-0 lead (Just about to hit submit and its now 3-0!!)

The other backs a goal for the second half.

So many league and so many bots its hard to keep track of so will work that out for you all.

Any questions let me know

[A Former User]

So this is now where I am at (No screenshots today as not on my server).

I have done some filtering on my Scalp FT bot and I have actually come across something that's made me now look at expending the time range but been specific with the League and price. I will explain more on Friday.

Second Half Scanner works very well ATM but needs some fine tuning but I am actually happy for this to go into Live mode on Saturday. Its stayed in the green during the months testing and i have reset it several times to start from scratch and always ends up in the green so to me if its making green its ready to test in a live mode (You can test and test in Practise mode but at some point it needs to go into a live world with live examples - I have tried dogs and Nags in practise and came out with riches only to find it dive bombed in live mode!!).

I have another bot ready to go on Friday as its specifically for the A-League and might as well catch the Fridays game with it.

I have had several other bots running in the background that i am happy with. I have not spoke about them as they are very complicated and have required multiple bots working with each other to work at one point i nearly chucked in the towel.

I will explain a belief into to the bots going Live on Friday .

What I thought I would do is show everyone the A-League Bot.How I got to where it is and how i put it together in a tutorial type post (it wont be 100% as i would like people to also start thinking a little more around what they are looking for ect)

Anyway that's me for now!

[A Former User]

Majour test today with my bots - I have around 30 going on tody all with the same shg bet/trade but it all depends on the league which it selects, the position of the team and how many shot they have seen in the first half.

It was a headake to put together but i think i have done it but will need testing today. I have paper traded 3 leagues and its $$ maker and a great steady earner.

Second Half scanner and FT scalp will take the test today!

Scalp FT needs filtering out as I have seen a few things such as games over 3.5 that produces more goals after 90min. But need to get to 1000 there really

Not much to report here atm but when Feb hits and we go live it should be interesting and will post bits about what I have noticed and learnt and spotted in my research ect

[A Former User]

So bit of an update.

Benn in London with work all week so not got much done and not been able to update to much but I have done 2 big changes.

Following on from some testing research from the bots I have running I have produced a bot that is doing rather well!

The Second Half bot - slight tweaking needed but looks a nice earner if crry on the way it does.

Scalp FT has Hit red but I will keep this one going to 1000 trades so i can look at some stats - i am pretty sure 0-0, 1-1, 0-1 and 0-1 leading into last 3min or so are dangerous as teams can sneak a win or a draw still!

The other thing I have done is looking to re-jig my data a little as I had a brain wave on the train. I am currently catagorising teams from each league depending on their position so, for example, the prem league...

I have 4 tiers that i am looking teams into... tier 1 the top seeking glory .. tier 2 those trying to still hunt down those glory sports and still have a chance. Tier 3 ... no hope for the top spot but still need to worry about the relegation area and Tier 4 ... the doomed.

By doing this then measuring the team's performance based on playing a team in that tier i should get some other stats ... for example

Man City stats playing a team from tier 4 such as Watford will bring different stats then it would if a team is playing two tear 1. The data is not going to be huge but i will have a few season worth and could play part of the full view!

Hope this gets done over the weekend

Following that its sticking some stats into live for feb then i move to stage 2!!!

Match Modeling !!!

[Martin]

@jarratt-perkins brilliant!