Wednesday 5 February 2014

Reply to Steve's (Daily 25) comments.....

Now that I’ve completed the monthly review for January, I want to take this opportunity to respond to a couple of comments that Steve made on his Daily 25 blog this week.  It’s always tricky writing a post in response to a little bit of criticism but I really want this to be a constructive post and hopefully Steve takes it in the way it was intended. I’m sure he will!

Steve’s first season with TFA was last season when he played a portfolio of systems and then this season, he refreshed his portfolio of systems he follows and so far this season, it’s safe to say that his portfolio has been nothing short of a nightmare. Here is Steve’s latest comment in full:

TFA is now really starting to frustrate me, Even though the higher level systems are about break-even, I believe most of the subscribers would be using the combined systems as they are the ones that have historically produced the best results. I believe having so many systems is a bit of a cop-out at times and saw it with Formbet as well. With so many systems, one is always going to be winning and Formbet goes to great lengths to let people know of any one system that is ahead, but overall the system has a massive loss. TFA isn’t so blatant in regards to this but does let people know that it isn’t up to him if they win or lose. This is the beauty of these sort of systems for the tipster, they are always going to have a stream of players winning and letting others know how great they are, and the losers just leave. I have a lot of respect for tipsters like Skeeve and FB Elite who back themselves completely. There is no hiding behind anything in what they do. I guess my real frustration lies in how bad I want The Football Analyst to succeed. It has high turnover at long odds and the potential to add a massive amount of profit per year, but the fact that I am down after almost 2 full seasons is not a good sign. I really do not want to drop the service, but at -$40,000 so far, it is not looking good.

I think before I get around to picking up on what Steve has said, I want to take a step back to look at the results Steve has had with the service thus far.

In his first season, Steve followed the 5 top Est combined Systems (6-22 thru to 8-22) and the 5 top Euro systems (E1-E7 thru to E3-E8). To quote Steve’s comment from his season review “I haphazardly choose 10 systems to play based on past performance, the amount of bets and a set goal I wanted to achieve. Everyone following TFA will have different portfolios and therefore different profits at the end of the season.”

In terms of results, Steve ended up with a profit of just over $35k last season.

“Total cost of the service was £150($228) and profit was $35,571.42. So the ending profit was $35,343.42. Even though it is more then 50% less then I had anticipated, I can’t be unhappy with such a large profit.
In total $587,115 was bet which equates to a ROI of 6%. Breaking it down further, the Established systems had a total bet amount of $328,623 and a profit of $33,876.01, which is a ROI of 10.3%. The Euro system had a total bet amount of $258,492 and a profit of $1,695.41 for a ROI of 0.65%.”

If we look at the overall results for all TFA systems last season, the results were as follows:


 What I’ve done here is simply rank the results of last season’s systems by amount of profit won. I’ve highlighted the 10 systems Steve chose to form his TFA portfolio of bets last season.

The best system Steve followed was ranked 19th in terms of performance at season end. The 10 systems Steve had as part of his portfolio were all ranked in the 46% percentile or lower. Quite difficult to do I suspect as if you were guessing, you would hope to hit a better portfolio.

The fact Steve made $35k profit wasn’t down to his portfolio selection. Basically, last season, when 33 of the 41 systems made profits, it was impossible to not make a profit unless you did something crazy but at the end of the day, Steve should have done better than $35k given his staking amount I think.

If we roll on to this season then, Steve put what he learnt into action and IMO, did the right thing by increasing his portfolio of systems and he went from 10 systems to 13 systems. How has his portfolio selection done this season?

This is from the latest blog post:

“TFA $439,126.00 $(39,161.08) -8.91%”  Based on Steve’s £75k bank assigned to TFA, he has lost 52.2% of his bank for TFA.

Steve doesn’t bother splitting the results on the blog now given he’s following so many systems (can’t blame him!) but similar to what I showed above for last season, here are the results of all systems to the end of January for all 52 systems (same 41 as last year plus 11 Draw systems)


 The first point to make is that only 35 of the 52 systems are in profit this season, therefore, you don’t need to do anything too crazy to make a loss this season. Even so, if randomly choosing 13 systems, you would still have a decent chance of a profit this season by guessing which systems to follow. 

As you will see, I again have highlighted Steve’s portfolio of systems for this season.  Steve has no systems in the top 1/3 of systems this season and in the second 1/3, he only has 5 systems with the other 8 systems in the bottom 1/3. In total, only 6 of Steve’s 13 systems are in profit even though 35 of the 52 systems are in profit!

Now, can you see a similarity to last season? Last season, I think Steve got lucky in the sense that given it was impossible to lose, he had to make a good profit but as always, you probably only get lucky like that once and this season, when the systems haven’t done quite as good, Steve is looking at a substantial loss.

Putting the system selection issue to one side for the moment, I think the other issue Steve has had this season is that he was forced to change staking plan after a poor September for the systems as quite simply, Steve was over-leveraged and didn’t have the Bank needed to play the portfolio he chose to play.

Again, this is maybe bad luck but then again, it’s the sort of mistake that can be avoided. Steve was aware that this 13 system portfolio was massively leveraged already and therefore, by playing too high a stake relative to his Bank, he was taking a massive risk and ultimately, it has come back to bite him.  Steve’s results aren’t a real reflection of the system’s performance and ultimately, the staking plan is to blame for this.

OK, getting back to system selection. Last season, Steve got a little lucky by following 5 correlated UK systems and 5 correlated Euro systems. This season, he chose the same 5 UK systems (6-22 thru to 8-22) and therefore, these systems were massively correlated. As you will see on the table, 7-22 thru to 8-22 are sitting 45th, 46th, 49th and 51st. Hence, although Steve thought he was following 4 systems, in reality, with the correlation of these, he wasn’t. Unfortunately, this has caused a problem this season. So, Steve is following the 4 worst Est Combined systems.

In terms of the Draw Systems, Steve took a punt on two combined draw systems this season. Not a bad idea and Steve ultimately settled on the top two combined systems since these have the lowest number of bets and I guess he learnt from the Euro systems last season!

Looking at the table, the top 4 combined systems for draws are sitting 5th,6th,9th and 10th. Steve’s systems are sitting 18th and 32nd. Again, Steve had a choice of 6 combined systems and has chosen the worst two systems to follow.

Looking at the Misc Systems, Steve had a choice of 6 Misc Systems. Steve chose TOX, STOY and STOZ. The other 3 Misc systems are sitting 2nd, 8th and 28th. Steve’s 3 systems are sitting 25th, 35th and 36th. Hence, Steve has chosen 3 of the worst systems in this group to follow.

Lastly, looking at the New Systems, Steve chose 3 of these 6 combined systems. The other 3 combined systems are sitting 1st, 12th and 23rd. Steve’s 3 systems are sitting 30th, 38th and 41st.

Steve has chosen the 3 worst New combined Systems to follow.

Putting the above together then, Steve took what he had in the first season and ultimately, made it worse this season! He has consistently chosen the worst systems to follow from each set of systems and it’s actually amazing that he has built what could be the worst possible portfolio from all of these systems I’ve seen this season!

If Steve was simply guessing, he couldn’t do as bad as this I think.

I guess the question that needs answered is….how much of this is down to bad luck or is it the case Steve could have learnt from last season?

Before I give my view on this, I want to highlight an experiment I have been doing this season in The Secret Betting Club forum. Given the issues some people have with building portfolios of TFA systems, I think there is maybe a demand for me to just offer ‘off the shelve’ portfolios for people to follow in future to make things easier. So, as an experiment this season, I built 10 sample portfolios so I can track how these 10 do.  Obviously, the more systems in a portfolio, the more diverse the portfolio can be in terms of types of systems followed. All portfolios are 1pt level stakes.

Below are the results so far this season which are tracked in the SBC forum:


From the 10 portfolios, 6 are in profit with 4 showing a loss. The worst portfolio (which was only 3 systems) has lost 20.6% of its betting bank this season. What the table helps show though is that if you follow a well diversified portfolio (in this case, 7 systems max), then you should be doing OK this season. The fact Steve follows 13 systems and is losing 50% of his betting bank points out that he’s not exactly following a well diversified portfolio this season and something has gone wrong with his thinking about how to use these systems!

So, getting back to the question I posed above, is it bad luck that Steve has ended up following a stinker of a portfolio this season? Part of me says yes but the other part of me says no. I follow 20 systems in my TFA portfolio. I have 11 (yes, 11!) of the same systems as Steve in my portfolio of 20. Given I’m in profit this season (not by far admittedly but a profit is a profit!), then it shows how unlucky Steve has been this season. Then again, if I was building a 13 system portfolio myself, I wouldn’t have 11 of the 13 systems Steve has in his portfolio!

Now, some reading this may be thinking that you need to follow lots of systems in a portfolio but that’s rubbish. Look at the SBC experiment results. Two portfolios with 5 systems are having an OK season with ROC of 30% and yet, they only have 5 systems. The key is, the 5 systems are well diversified and a mix of systems and therefore, although the small portfolio has a losing system (one on each set of 5), the other 4 systems are doing OK.

I spoke all last Summer about how diversification is going to be the key going forward for people following TFA who want to make profits without carrying a crazy amount of risk and yet, some people continued to choose highly correlated portfolios to follow in the hope that variance doesn’t strike and they make huge profits. Steve isn’t alone (I know a fair number of others as I pulled together portfolio sheets for them in the Summer) but like I’ve always says, I supply the system bets, what you do with them is up to you!

So, I do think Steve has been very unlucky this season but then again, I could say he was unlucky last season as he didn’t have a great portfolio then either. I keep saying this over and over but you make your own luck at this game. Some have been doing much better than me this season but that’s because they plumped for other systems which I would have deemed higher risk but they have been rewarded. This game is all about risk and reward and those people in profit this season deserve to be in profit. As for those making a loss……they put themselves in that position with their portfolio unfortunately.

Just to be clear, none of what I’m saying detracts from the fact that the systems have had a poor season so far. You can see that in the results above for this season compared to last season. The fact I’m only slightly better than break-even gives an indication of how tough this season has been for me, so if I’m not making a profit, not sure I expect anyone else to be making a profit. However, what I don’t expect and what shouldn’t have happened is people should not be losing 50% of betting banks or going broke and yet, I’ve read comments on Steve’s blog this season from others saying similar things!

Right, getting back to the point of this post, the reply to some comments Steve made.  

“TFA is now really starting to frustrate me” – Anyone who reads my blog can surely see how frustrated I’ve been all season myself. A fair comment and one I agree with!

“Even though the higher level systems are about break-even, I believe most of the subscribers would be using the combined systems as they are the ones that have historically produced the best results” – This is the first comment I have an issue with. If we look back at the results I’ve posted above for last season, let’s look at some of the system results. The top system this season is system 31-41, was 8th best last season. The 2nd best system this season is 6-21-31, it was 16th last season. The 3rd best system this season is 31, was the 1st best last season. Ignoring the 4 draw systems which fill 4th-7th this season, the 8th best this season is system 21-31 and it was 11th last season.

I’m not wanting to rub salt into the wounds here but these 4 systems were not that difficult to find. Maybe not too surprising to know that system 31 appears in most of the SBC experiment portfolios and in my own portfolio whereas system 8-21 for example doesn’t appear in any of the 10 SBC portfolios.

I get what Steve is saying with his comment but I think he needs to open us his mind to how we can best use these systems to make money. Following highly correlated systems isn’t actually the way to make profits and reducing your risk at the same time. It actually increases your risk IMO and when variance hits, it can kill your portfolio (as has happened to Steve this season).

“I believe having so many systems is a bit of a cop-out at times and saw it with Formbet as well. With so many systems, one is always going to be winning and Formbet goes to great lengths to let people know of any one system that is ahead, but overall the system has a massive loss” – Not surprisingly, I have a massive issue with this comment! I always report all results for all systems and even if a set of systems bomb (like the Euro systems last season), they remain part of the live results for the systems and always will be. Same as when the Under/Over systems retire this season. The results won’t suddenly vanish, I proofed these bets and therefore, they are part of the service returns.

I never play on the fact that any system is winning more than any other and anyone who reads my blog results updates will see this.  If you read the monthly review, I think I make it clear that I don’t think anyone has made a profit in January (I know some have as I saw it on Twitter!) and therefore, I’m hardly playing up my system returns.

In addition, given I follow my own portfolio of bets, my mood on the blog has hardly ever been upbeat this season given I’ve been losing money for the most part, so I don’t think having so many systems is a cop out at all. If I ever start reporting on the top 10 systems or something crazy like that, then shoot me!

“TFA isn’t so blatant in regards to this but does let people know that it isn’t up to him if they win or lose. This is the beauty of these sort of systems for the tipster, they are always going to have a stream of players winning and letting others know how great they are, and the losers just leave.” – I have very little turnover of members as Steve is aware and if I wanted to have a stream of numpties joining TFA just to get subscription fees, I don’t think I’d be closing the doors ever Summer and turning people away. If I was really not worried about the people losing with the service, I wouldn’t be writing this blog post to draw publicity to Steve’s losses. A poor comment IMO although I reckon it was written in the heat of the moment.

“I have a lot of respect for tipsters like Skeeve and FB Elite who back themselves completely. There is no hiding behind anything in what they do.” Errr… I have probably the most public blog from any well known tipster, I update my results weekly on the blog, I never hide any results for any systems, I publish my monthly reviews on here, I have a Twitter account where people constantly discuss my results with me, all my bets are proofed by the SBC and reported on by them. As for backing myself, I put my money where my mouth is and although I may not have the bankroll of Steve or Skeeve, I’m at a level now whereby the money I win or lose matters to me, more than I like to admit at times. Again, a poor comment IMO.

“I guess my real frustration lies in how bad I want The Football Analyst to succeed. It has high turnover at long odds and the potential to add a massive amount of profit per year, but the fact that I am down after almost 2 full seasons is not a good sign. I really do not want to drop the service, but at -$40,000 so far, it is not looking good.” – I think this ultimately lies at the root of Steve’s comments if I’m honest. Steve knows I’m not trying to fleece anyone here and therefore, he wants me to succeed as do the majority of my subscriber base. He knows the work and effort that goes into TFA from me and everyone following and therefore, when things are going well, there is no better feeling in the world for me and for anyone following I think.

I think the issue isn’t so much this in isolation but last season too. It’s criminal that Steve’s down over 1.6 season’s following TFA but I’ve no idea how he only made 50% ROC last season and it’s even more puzzling that he’s lost over 50% this season!

If I was in Steve’s shoes and simply looking at the returns he’s achieved, I think he’s got every right to be considering dropping TFA. Making no money over this period of time is unacceptable and the ups and downs he’s had probably make it seem worse than it is. However, so much of this is linked to portfolio selection and when you follow a large set of systems that are correlated and overstake which then leads to you having to reduce staking after a poor period, then I think this is to blame almost as much of the systems.

As always, the above is my view and I’m sure some reading this may think I’m being a little harsh in places but as always, it’s how I feel. I personally think Steve’s a really good guy and I really like reading his blog, even when it’s advertising some massive losses from following my systems! However, I thought some of the comments on the last post were a little harsh and I’ve taken this opportunity to reply. 


My hope is that the systems Steve follows turn around and do better in the next few months but as always, I’ll continue to report back on all systems. Steve is one member of a subscription base and ultimately, he’s maybe now my most ‘famous’ member which doesn’t look great for the service but that’s life. If anything, Steve’s portfolio shows what can happen if you get a little unlucky and maybe follow a portfolio that’s not as well diversified as it should be. However, I don’t think Steve’s losses detract from the service too much if you look at the overall results over the last 2 seasons. 

8 comments:

  1. Interesting read, at this point I am down 42.3 units and not happy at all, not happy with myself!

    I'm following just two systems, 21 and STOZ, before the season I did a fair amount of research and decided just to follow those systems in full, however come the first day I read one of your blog posts and somehow gathered from that it would be a good idea to exclude a whole lot of leagues based on past performance. Of course those leagues, i.e. league two were the profitable ones!

    Obviously those systems have under performed but I can only blame myself for digging this hole, I think steve needs to learn from this season much like I will as he has certainly made a few mistakes.

    ReplyDelete
  2. Hi Anon.

    Sorry to hear about your losses so far. Not great news.

    I think I’m happy to hold my hands up regarding LG 2 Aways and maybe I should have been stronger in my conclusions to say that it’s likely to only be variance and to keep following a set of bets that have been heavily loss making over the first 3 seasons but it’s not always easy.

    I said in the Summer I would give them another season and then I would maybe make a decision about whether or not my ratings were cut-out for these bets. I had no explanations for why Aways were doing so badly in this league but so well in all other leagues for the first season. Likewise, I have very little explanation about why League Two Aways have done so well this season when Aways in every other league are really struggling.

    With hindsight, cherry picking data is something we’re all prone to do but I think we have to keep coming back to the premise of the way the ratings work. There is no rational explanation for why Lg 2 aways were so bad over the first 3 seasons and therefore, without an explanation, I’m not sure dropping them makes any logical sense. That was the conclusion I kept coming back to myself and although not bullet proof reasoning, I had no reason to believe things wouldn’t turn one day!

    Yes, dropping league two aways would have increased profits over the first 3 seasons but I guess the question is, what happens if they come good over a season and although a longshot, what happens if the other aways struggle? At the start of the season, this probably had a very small chance of occurring but unfortunately, as we sit here today, it has happened.

    I guess to be fair to my systems and me, my results are always shown including League Two aways historically which has brought down my results a great deal and therefore, I shouldn’t apologise for the fact that they have saved the returns to an extent this season.

    What I can apologise for is the poor performance of the other bets apart from Lg 2 aways. I sometimes read comments on other blogs or sites regarding TFA and I sometimes struggle to recognise what they say about me or the systems. I’m fed up admitting in blog posts and emails about how shi* this season has been and I’m always amazed when people say stuff like I’m making the results seem better than they are or I have so many systems to hide the results of the systems.

    Do people realise how stupid that sounds? If I didn’t have an edge, I wouldn’t be running 52 systems as it’s highly likely if I had no edge, I wouldn’t fluke a return over 10,000 bets a season!

    Anyway, apologies for the returns this season you’ve suffered. I can’t hide from the fact that this season has been very poor for the service so far and unfortunately, for those that have either increased staking off the back of the last 3 season’s results or those that only joined last Summer, I’m sure it is a bit of a shock. I spoke about this earlier this season but given the results over the first 3 systems and the high % of systems which made profits, people expected to join the service, pick any way to play the bets and thought it was impossible to lose. That was the case for the first 3 seasons but as always, when it got tough, it was always going to be the case that those who had weaknesses in their strategies to follow the service would suffer the most. However, they do tend to benefit the most when it goes well!

    Anyway, thanks for the comment and good luck for the rest of the season.

    Graeme

    ReplyDelete
  3. To me it looks like Steve is covering all the different algorithms.
    Which systems are in those two 5 system portfolios at SBC? I want to see what you mean with a diverse portfolio.

    I am not a member but I plan to be in the future. For a new customer it is quite a mess with all these systems. I do hope you just simplify it a bit and offer some ready-made diversified portfolios like you are testing now with SBC.


    ReplyDelete
  4. Hi Anon.

    Steve is following systems from the Est Systems (5), New Systems (3), Misc Systems (3) and Draws (2). Nice approach to vary the number of systems by the different sets of system and a strategy I adopted myself this season, so no issue with that at all.

    However, the first issue comes because algorithm 1 (systems 6,7,8) hasn’t had a good season. Steve follows 6-22, 7-21, 7-22, 8-21 and 8-22. Hence, if system 6 has a poor season (like is happening), he’s screwed with these 5 systems as they all rely on 6 (7 and 8 are a subset of 6).

    For the New Systems, 2 of his systems are 33-41 and 33-42 which only really rely on 33. 33 has less than 100 bets a season, so is prone to variance. 31-42 is a nice system to follow IMO, regardless of how it is struggling this season.

    Misc Systems, 6 to choose from. 2 sets of systems really (6-32, 21-31 and 6-21-31) and TOX, STOY and STOZ. Steve follows the last 3 which are all heavily correlated as I’ve never tried to hide. Hence, if one does bad, they all do bad.

    For draws, Steve follows the top 2 draw systems which have less than 50 bets a season. Again, variance can take a hold if things don’t go well.

    I said it in my post, Steve has been unlucky in an extreme way. This season, the bets that appear on the most systems have really been disappointing and therefore, I’m not saying anyone should be making a profit. I think break-even is a good result this season and a small loss is acceptable. What isn’t acceptable is people losing 50%+ of betting banks as I don’t see how this should happen when no individual system has lost more than 25% of its betting bank at any time this season. Clearly, something is going wrong when people are losing so much when the systems individually aren’t doing that bad.

    When 66% of the systems are in profit at the end of January, if randomly guessing a portfolio, you are odds on to make a profit. To lose 50%+ of a bank when 66% of systems are in profit isn’t right. Yes, it can happen (has happened to Steve!) but something fundamental must be wrong with the approach. The purpose of having so many systems and algorithms is to reduce risk, not maximise profits. If you want to have a high chance of a profit, follow one system and hope you hit a winning system! 66% chance at the end of Jan!

    ......

    ReplyDelete
  5. Anyway, I’m sure there will be a few changes this Summer and as I said in the SBC forum, I realised before this season started (hence why I put that thread up there before the season started!) that if the service ever suffers a poor season, I expect those who build their own portfolios which are correlated or data mine are going to lose badly. That’s what I’m seeing.

    IMO, this game is all about controlling risk and exposure. Profits come easy when the systems do well but the downside risk is what your portfolio needs to protect you from.

    In terms of the SBC portfolios, it’s easy. 5 systems, one Est System, one New System, one Draw System, one Euro System and one Misc System. As I said in the post, building a diversified portfolio isn’t too difficult, you just need to give it a little thought! The highest combined system is 32-41. Why? Well, a 5 system portfolio shouldn’t be centred on the higher combined systems as they don’t have enough bets and are prone to variance. If you are running a small portfolio, you need a big sample of bets that are largely uncorrelated.

    I know TFA isn’t for everyone and therefore, anyone who can’t see how the service works and gets confused by the systems shouldn’t really be joining the service!

    What I will probably do next season is have another way of following the systems whereby people can subscribe to a portfolio of bets. This at least puts TFA on the same footing as many other services and allows comparisons to be made more easily. Doesn’t reduce the risk of having a poor portfolio tbh, just means your fate is put into my hands. Given I seem to do OK each season (not as well as some during the good seasons but nowhere near as bad as some during the bad seasons!), then hopefully I can build some decent portfolios for people to follow.

    End of the day, I’ve always said that if I can’t make a profit betting, there won’t be a service. This hasn’t changed. I don’t want to be a tipster, I want to be a successful bettor. If I can’t make a profit, there will be no service even if others can make a profit.

    Cheers for the comment.

    Graeme

    ReplyDelete
  6. I am following 19 UK-systems with a 150 point bank (should be 200, but I took a higher risk). I am staking 1 point on homes, and 0,5 point on aways and draws:

    6-21, 6-22, 7-21, 7-22, 8-21, 8-22
    31-41, 31-42, 32-41, 32-42, 33-41, 33-42
    TOX, STOY, STOZ
    D1-D6, D1-D7, D2-D6, D2-D7, D3-D6, D3-D7

    I follow all the same systems as Steve do, plus 6 more. The difference in result is huge! So far my results are +10 pt. The lowest point early in the season was -100pt and the highest point was in early january with +50pt.

    Have I chosen a better overall portfolio then Steve, or have I just been lucky with my choice of systems? What do you think Graeme?

    Best regards,
    Vidar

    ReplyDelete
  7. Hi Vidar.

    I think it’s 21 systems you must be following rather than 19. :)

    I’ll respond to this in a post as people won’t see my response on this post since it’s from a wee while back.

    Cheers,

    Graeme

    PS. Well done to be in profit at the moment! Any form of profit this season is amazing given the results.

    ReplyDelete
    Replies
    1. Yes, 21 systems. Don´t know how I added that up to 19 when writing this post. Looking forward to your post :-)

      Thanks,
      Vidar

      Delete