Tuesday 3 May 2011

Some Thoughts.....

OK, to follow on from my last post about the weekend results.

I deliberately took the last couple of days away from things to do with the footie. Anyone who knows me or who has been following the footie bets or even just reading the blog this season knows how much time, effort and dedication has gone into this project over the last 16 months and therefore, even though February was hard to take (it was my first ever losing month on the footie personally), April has been much harder to take.

I don’t want to bore everyone on here or in the TFA forum about how much I’ve put into this (it’s not hard to see as this sort of thing eats up your time like you’d never believe) but simply, this takes up so much of my time, I really need to have complete confidence in where I’m going with it. If the day comes when I feel like I can’t make this work going forward, I need to pull the plug on it and walk away as I can’t justify my time on this without making money from it.

I’m not afraid to admit that I’ve now made next to no money on the footie this season (own fault really as I changed stakes a few times this season and at Christmas, I thought I’d maybe built the Holy Grail when it comes to footie which meant another increase in stakes!) but unfortunately, the second half of the season has nearly wiped out all my profits from this season. Putting my personal betting to one side, the question I need to ask myself continually with this project is, am I any closer to finding something that works?

Even after February, I had a massive belief that it was a one-off month and after a fair bit of work looking at the monthly results statistically, I was confident that February was a once in 20 month event. Hence, months like February would come along 5% of the time and therefore, I knew what to expect.

Anyone who’s been reading the blog for a while will know all about February and all the statistical testing I carried out to try to find out how common a month like February was based on the past month’s results. Using a simple Standard Deviation approach, I figured that February had a circa. 5% chance of happening based on past months. The draw % was 38.5% in this month and to put it in perspective, the next highest month was 33.5%. I took February on the chin and said to myself that if months like this happen once every 20 months, I can take it and not worry too much about it. It was part of the game.

March then came along, won back all the points lost in February and even though the ROI on the systems had taken a massive hit after breaking even in 2 months, I wasn’t overly concerned about what had happened and still had belief that I was on track to build something really special here.

That led us onto April…..

I’ll spoil the fun from my monthly review but the draw % for April is 42.1%. Now, if you had hit a 42% strike rate with draws over a small sample of bets, I can understand that things like this can happen. However, the number of bets in April has been 439 bets. Hence, I’ve hit 185 draws in 439 games.

I’ve done the same analysis as I did with February and simply, the chances of April occurring based on past month’s draws % lie between 1% and 2%. Hence, we’re looking at April happening maybe once every 50 months.

I can even go a step further than this and if I look at 2011 as a whole, the draw % is 36.1% from 1,606 bets. If I was aiming for draws, to hit a 36% strike rate over so many bets would be the best system I’ve probably ever seen on the football.

April alone has made 200pts profit from backing the draw and 2011 has a profit of 400pts from backing the draw. This is from the same 1,606 bets. A 25% ROI from backing the draw.

I’ve been thinking long and hard over the last 2 days about whether this is something that I should have maybe been more aware of but I don’t think I could have been. Before this season, backing the draw in every game has lost 1,800pts over 12,000 games. This season, the draw has made a profit of 226pts from 2,800 games. Clearly, the first half of this season has followed the past and made a loss but for some reason, the second half of this season looks nothing like the past.

I keep going over and over it in my head and thinking to myself, should I have spotted this could happen? Could I have saved myself all of the losses from Christmas? Could I have saved others some money too?

I think at the end of the day, I keep coming back to the same thing. February was a once in a 20 month event and April is a once in a 50 month event. The chances of both these months occurring within 3 months? Well, I don’t want to guess as I’m not sure if they are both independent of each other (20*50 if they are) but whatever way, we’re looking at 500/1+ I suspect.

Based on the evidence of 14,838 bets to the end of April 2011, it was impossible for me to imagine that February and April could happen together. Yeah, I could imagine either month could happen in a season but the chances of both happening within 3 months…..I couldn’t have imagined that happening.

Now that it has happened, where does it leave me for the future…..Is this project dead? My systems now look hopeless after their first live season.

I’ve given it a lot of thought over the past 2 days and I think it would be fairly easy for me to pull the plug on this and walk away. For a start, I’ve never received a penny from anyone for anything I’ve done this season and therefore, I’ve got nothing to feel bad for from that respect. Yeah, I’m disappointed some people may have lost money or not won the amount they thought they’d win at Christmas based on results to then but then again, I’m in the same boat. Considering I’ve done all the work, I should feel slightly more aggrieved I suspect than anyone else who has been following the footie!

Even though the easy decision is to walk away from this possibly, I feel like I owe it to myself to give this another shot next season. If anything, I want to prove to myself and anyone who’s been with me on this journey that April is a once in a 50 season event when it comes to hitting so many draws!

Take away February or April from my results and this season would be hailed as the best first season by a football system/tipster for a long-time. Unfortunately, I can’t do that but no one can take away from me the fact that at Christmas time, it looked like I had built something that could achieve a 20%+ ROI on the footie across a massive sample of games. January and March backed this up further but for whatever reason, the draws have killed me in February and April.

So, I’ll be giving this another go next season as I have a lot to prove I think. I’ve got lots of ideas around how I can use the draw to my advantage and looking at the long-term results of DNB and DC, you can see lots of evidence that I can probably afford to cover the draw in some games to try to reduce the variance of my returns for when months like February or April occur.

I’ve also got some ideas about how I can tweak the two rating algorithms I’ve got at the moment (especially systems 21-23) and I’m also going to consider building a 3rd algorithm that will look to maximise returns from DNB or DC instead of maximising returns from traditionally H/A betting.

I feel really down about this season at the moment and I’m disappointed about how it has went from Christmas but I honestly believe that at Christmas, this is the closest I’ve ever come to seeing something that I thought could actually work long-term when it comes to gambling. You don’t fluke 4 months results over 1,000 games and even though the season has went down the tubes, my ratings and systems have still been profitable this season overall if I look at the full season. That has to count for something!

Anyway, that’s my thoughts at the moment. I’ll get round to my monthly review sometime this week. I’ll mention it on here when I post it on the season’s results section.

A quick word on the recent comments from Your New Blog Reader. I think you raise some very good points and as the week goes on, I’ll get around to answering your questions or commenting on the points you make. It’s great to have a new reader tbh as it can get quite lonely at times on here writing to myself and I’d love some more interaction on here with anyone who is reading the blog. I’m sure some people have questions or comments to make, so feel free to post them up and I’ll respond!

2 comments:

  1. Tried to comment before but it didn't post for some reason.

    Just remember that 10% ROI over a season is an absolutely fantastic achievement. Many pros work on 5% or smaller margins, and make a whole heap of money doing it. You can't blame anyone for the staking changes (I believe any but the most hardened gamblers would have fallen into the same trap) - if I had to price up you having a winning system for next season, I wouldn't even offer 1.01, i'd be so confident from having read the blog (and also from looking at some of the teams you select, i'm 100% sure you are on the right side of the games at the bookies prices).

    Good luck, keep it real and in perspective (which you do very well).

    ReplyDelete
  2. Hi Adam.

    Thanks for your comment. You'll see I've added a link to your blog mate. I don't go out my way to add links to other blogs as I only add links to blogs I read myself and I'm looking forward to catching up with your blog as I hadn't seen it before you posted mate. I'll go back and take a read through it from day one as it looks interesting.

    Yeah, I totally get what you are saying about a 10% return from the footie but what has happened this season has meant that it's been difficult to make good money the way the returns have been achieved. Most people following the bets (including myself) started with smaller stakes and increased stakes as the season progressed but with hindsight, this wasn't the best idea for a system that is totally unproven and playing level stakes all season would have been the way to go. In addition, many people (myself included) have dabbled with different systems as the season progressed and I've increased the number of systems/bets since Christmas which again, wasn't the right thing to do looking at what has happened.

    When I set out, the 10% ROI target was my base minimum that I was looking for and now, I'm nowhere near that with my own betting. My ROI would be about 0.0000001% now!

    When I look at the level stakes returns from the systems and some returns I've made, I can only cringe really. I'm break-even on some systems that have a 10%+ return and I'm losing on systems that have a 5% return.

    I know from the TFA forum that many have made the same mistakes as me (hard to be too critical on others when the guy finding the bets has made a mess of things himself!) but this season was always meant to be a trial season and I think this was maybe lost around November/December when the profits were rolling in every week. I try my best to keep my feet on the ground along with others but I maybe should have played down the returns from September to December more as it was less than half a season. The fact the ratings were outperforming my backtested results meant I thought I had a fairly big way to go from making a huge profit to losing but as I've seen, it doesn't take much to turn wins into draws!

    The whole draw thing is muddying the waters since Christmas as I had no intention of going down the Asian Handicap route with these ratings but looking at the evidence in the context of the past few seasons now, you can argue that I should be playing DNB or DC on these bets to help smooth the returns. Something for me to work on this summer for sure.

    Thanks for the comment and at times like these, it is good to get a second opinion on how things are going from someone who's not too caught up in this bubble I'm in and everyone else is in who has been following the bets this season. Trying to keep perspective after losing a small fortune since Christmas is easier said than done!

    Cheers,

    Graeme

    ReplyDelete