FtWF; Update: Sep 13-14, 2008

There were three matches this past weekend, and all won. Results are as follows:

Sep 13, 2008 (Rangers v Kilmarnock)
Start Bank/Bet Amount: $125.00 / $25.00
Result/Odds: Win / 1.30
Profit/Loss: +$7.50
End Bank: $139.29

Sep 13, 2008 (Inter Milan v Catania Calcio)
Start Bank/Bet Amount: $125.00 / $25.00
Result/Odds: Win / 1.29
Profit/Loss: +$7.25
End Bank: $146.50

Sep 14, 2008 (Real Madrid v Numancia)
Start Bank/Bet Amount: $125.00 / $25.00
Result/Odds: Win / 1.27
Profit/Loss: +$6.75
End Bank: $153.25

So after a little bit of stumble, looks like we are on our way Next games are this weekend, there will be two qualifiers.

FtWF; Update: Aug 31, 2008

Another game done and time for another update. In my last post, I had indicated that there were two games this weekend, but I was incorrect; the Man Utd v Fulham is actually in Jun (I had made an error in entering the date in my spreadsheet) The only game this weekend was the Olympiacos v Atereras Tripolis. Here are the results:

Aug 31, 2008
Start Bank/Bet Amount: $125.00 / $25.00
Result/Odds: Win / 1.89
Profit/Loss: +$22.85
End Bank: $131.75

This gives me a $6.75 profit on the season thus far (just four games bet on)

Next bet will be Sep 8 with Inter Milan v Catania Calcio. Stay tuned…

FtWF; Update: Aug 26, 2008

There was only one game to bet on over the weekend: FC Porto vs Belenenses. Results are as follows:

Aug 24, 2008
Start Bank/Bet Amount: $125.00 / $25.00
Result/Odds: Win / 1.31
Profit/Loss: +$7.75
End Bank: $117.25

There are two more bets for this weekend: Manchester United vs Fulham and Olympiacos vs Asteras Tripolis. I will be posting more updates then.

FtWF; Update: Aug 18, 2008

Sorry I have not updated this blog in a few days; I have been away attending my brother’s wedding out in Ontario.

While I was away, the English and Scottish Premier leagues started up, and I had pre-bet two games on it using the “Four The Win Football” system:

• Aug 16: Rangers vs Hearts
• Aug 17: ManU vs Newcastle

The Rangers won while ManU tied. Since i am flat-betting, I will not be increasing my bet size on the next game, which will be Porto FC vs Belenenses on Aug 24th.

Results for the weekend are:

Aug 16, 2008
Start Bank/Bet Amount: $125.00 / $25.00
Result/Odds: Win / 1.38
Profit/Loss: +$9.50
End Bank: $134.50

Aug 17, 2008
Start Bank/Bet Amount: $109.50 / $25.00
Result/Odds: Win / 1.38
Profit/Loss: -$25.00
End Bank: $109.50

Just as a reminder, I am going to flat bet the season, not increasing my bets, to give a more true test of the system. I will also be tracking results using two other betting systems, but they will be theoretical, and not actually bet. I may make my spreadsheet available periodically to show the results.

Review: European Football System: FTWF, part 2

OK, in all future posts, I will refer to this particular Football system as “FTWF”, which is short for “Four The Win Football”.

As promised, I have done some “Risk of Ruin” calculations to determine what a good bankroll size would be to do this system with. As you may recall from my previous post on this subject, I initially did some calculations using 20% of the bank as my betting unit, but felt that without further data, it might be safer to use 10% of the bank as the betting unit.

To do the calculations, I used the the “Bankroll” calculator at this site. While the site is really for calculating a poker bankroll, it is still applicable here.

Using my spreadsheet, here are the results:

Flat Betting
Betting Unit: $25
Average Profit per Game: $3.96
Standard Deviation of the profit (using the “STDEV” function of the spreadsheet): 13.71

Plugging the numbers into the “Bankroll” calculator like this:

Win Rate: 3.93 (this is is average profit per game)
Standard Deviation: 13.71 (the standard deviation of the profit/loss)
Risk of Ruin %: (see below)

Using a “Risk of Ruin %” of 5 resulted a a bankroll requirement of only $71.64; “risk of Ruin %” of 1 resulted a bankroll requirement of $110.13

1-2-3-1 Staking Plan
Betting Unit: $25
Average Profit per Game: $4.62
Standard Deviation of the profit (using the “STDEV” function of the spreadsheet): 16.05

Plugging the numbers into the “Bankroll” calculator like this:

Win Rate: 4.62 (this is is average profit per game)
Standard Deviation: 16.05 (the standard deviation of the profit/loss)
Risk of Ruin %: (see below)

Using a “Risk of Ruin %” of 5 resulted a a bankroll requirement of only $85.32; “risk of Ruin %” of 1 resulted a bankroll requirement of $128.39

Increasing Betting Unit Based on 10% of Total Bankroll
This one was a little more tricky, since I needed to start with *some* bankroll number as a basis for the calculation. To arrive a betting unit of $25, I needed to assume a bank of $250. Here are the results:

Betting Unit: $25
Average Profit per Game: $4.70
Standard Deviation of the profit (using the “STDEV” function of the spreadsheet): 22.80

Plugging the numbers into the “Bankroll” calculator like this:

Win Rate: 4.70 (this is is average profit per game)
Standard Deviation: 22.80 (the standard deviation of the profit/loss)
Risk of Ruin %: (see below)

Using a “Risk of Ruin %” of 5 resulted a a bankroll requirement of only $165.67; “risk of Ruin %” of 1 resulted a bankroll requirement of $254.68

Interestingly, if I only wanted a 1% “Risk of Ruin”, then my “assumption” of using a starting bank of $250 for this one was actually very close. Actually, looking at all the results, my initial instinct of being able to use 20% of your bank as a betting unit seems to be a good one, but I am still going to use 10% of my bank as my betting unit until i get further data (two or three more seasons)

A couple final calculations:

1-2-3-1 Staking Plan: Each Team Bet At Its Own Series”
Betting Unit: $25
Average Profit per Game: $6.90
Standard Deviation of the profit (using the “STDEV” function of the spreadsheet): 23.42

Plugging the numbers into the “Bankroll” calculator like this:

Win Rate: 6.90 (this is is average profit per game)
Standard Deviation: 23.41 (the standard deviation of the profit/loss)
Risk of Ruin %: (see below)

Using a “Risk of Ruin %” of 5 resulted a a bankroll requirement of only $118.97; “risk of Ruin %” of 1 resulted a bankroll requirement of $182.88

Flat Bet: Each Team Bet As It’s Own Series
Betting Unit: $25
Average Profit per Game: 5.31
Standard Deviation of the profit (using the “STDEV” function of the spreadsheet): 18.97

Plugging the numbers into the “Bankroll” calculator like this:

Win Rate: 5.31 (this is is average profit per game)
Standard Deviation: 18.97 (the standard deviation of the profit/loss)
Risk of Ruin %: (see below)

Using a “Risk of Ruin %” of 5 resulted a a bankroll requirement of only $101.51; “risk of Ruin %” of 1 resulted a bankroll requirement of $156.05

So there you have it; pick which betting strategy you want to adopt, and size your bankroll appropriately.

One final question you might have is what is more appropriate: 5% Risk of Ruin or 1% (or even lower??) That is really up to you. If you are doing this as a hobby or secondary income, then you are probably safe with using the 5% figure, but if you are doing it for a living or just don’t want to go broke, then 1% (or lower) would be more appropriate.

I hope this helps you. As always, if you have any comments on this article or the numbers used in it, please feel free to comment.

Review: European Football (aka “Soccer”) System

Wow, two posts in one day

I was surfing around for some good Betfair betting/trading systems, and have found some good ones, which I will cover in some future posts, but wanted to share my thoughts on a Football system that I came across at the Four The Win Football site. It’s a free system (well, it’s not really a system, as you will see), though they have a couple other systems, notably horse racing, that they sell (and which have had really good reviews, but are beyond the scope of this post)

Anyway, since it is free, I will share the “system” here with you.

It involves betting on the top-tier teams in European Football who had the best home records (no more than three, maybe four, non-winning results the previous season; so in other words, no more than 3 or 4 draws or losses) You then bet on those teams the following season when they play at home.

Note: The results posted on the website report in “Great British Pounds”, but since I do not have the symbol for “GBP”, I am going to use exactly the same numbers, but use “$” instead, as I do not believe that changing the monetary symbol actually changes anything.

They posted their results for the previous season, which results in about $21,160 over the course of the 10-month season (Aug-May), betting $1,000 each game. They recommend the following “staking” pan: 1-2-3-1, meaning that you bet one unit, if you lose, your next bet is two units, then three if that one loses, and if there’s a 3rd loss, you just cut your losses and go back to one unit.

I put their results into a spreadsheet and played around with things a little bit (trying different betting strategies and the like). Here are some interesting things I found:

1. I played around with the best size of bet to bankroll ratio, and while I found that making your “one unit” bet 20% of your bankroll worked, until further data is available, I would probably be more comfortable with using 10% of your bankroll as your “one unit”. I will probably do a “risk of Ruin” calculation and see what the best bet size/bankroll is based on that, and post an update to this in the very near future.
2. Their results are missing four final games for the teams they bet on. I do not know if the omissions were on purpose, accidental, or if they just simply did not bet on those games. Adding in those “missing” results changes their overall profit from $21,160 to $19,880, which is still quite a hefty profit, although adding in those “missing” results actually resulted in an overall loss on one of the teams. Note that since I did not have the odds for those four “missing” results, I averaged out the odds from all the others games which I used in my calculations for those “missing” games (the average odds worked out to 1.36)
3. The results they post list how much money was made on each team, so they seem to treat the 1-2-3-1 staking plan on a per-team basis. Betting per-team, out of the 95 bets, it was necessary to go to the “2″ bet 11 times and the “3″ bet once.
4. Playing with things a bit, I had a look to see what would happen if, instead, you mixed everything up and just bet on each team as their games came up on the schedule, and did the 1-2-3-1 staking plan that way. The results were quite a bit different:a.) $17,560 profit (a difference of -$2,320, using the “adjusted” figure of $19,880 above). While a bit less, it’s still a pretty healthy profit considering the “blue chip” teams your are “investing” in
   b.) Out of the 95 bets, it was necessary to go to the “2″ bet 12 times and the “3″ bet twice. Actually, one of the “3-bets” were the last three games of the season, and the last game “won”, though even if it had lost, the profit on the season would still would have been a respectable $13,480
5. I played around with one other betting strategy which involved having your base bet as 10% of your overall bank but staying at that level until your bank reached certain “milestones”. For example, starting with a $150 bank, you would bet $15 (using the 1-2-3-1 staking plan) until your bank reached $200, then you would start betting $20 until you reach $250, then you would bet $25 and so on. The “milestone” amounts were set at $150, $200, $250, $500, $750, $1000, $1250, $1500, 1750, $2000, $2500, $5000, and ending at $10000. Using the above results, starting with a $150 bank and $15 base bet, and betting on each team as their games came up in the schedule would have resulted in a profit of $446.80 with the base bet increasing to $50.
6. Finally, as a true test of the system, I tested to see how it would have done by flat-betting, which of course tells us truly whether or not a system is profitable. I tested flat-betting using a $10,000 bank and $1,000 as the “base” bet (i.e., treating each team as it’s own “series”, and mixing them all together) so the results are easy to compare to theirs. This resulted in the following:
   a.) Each team as its own “series”: +$15,050
   b.) All teams mixed: +$14,150

All in all, it would appear that if you were not planning on increasing your “base” bet size as your bank increases in size during the season, then betting each team as it’s own “series” would likely yield the best result, but no matter what method of betting you decide on, you would have been looking at similar results.

I will post an update in the very near future with the results of a “Risk of Ruin” calculation, including how I derived at the answer in case anyone has any further insights. In addition, I plan on using this system on the upcoming season, then the increasing stakes through the season as my bank increases. I will post the results (good and bad) as it progresses.