Entries Tagged 'Footy' ↓

Remarkably good results this week: “money” got seven matches right (it missed, and did so spectacularly, the Lions-Bulldogs match), “home” got six and “ladder” got five. One interesting fact is that the bookies were giving a 50% chance on either result on the Geelong-St. Kilda game; my “tie-break” method in the money strategy was to go for the home team, Geelong (now, would someone tell me just why Geelong was the home team at the Telstra Dome? Geelong is at least 60km farther away from the Dome than St. Kilda!); Geelong won.

Financially, all three strategies were well into positive territory this week, with “ladder” doing the best of all due to, in large part, the strangely skewed odds payed for the already mentioned Lions-Bulldogs game. After seven (of 22) rounds, “money” is still returning a net loss, but the others have reasonable gains.

In any case, here are the numbers:

Links:

• Round 6 (previous)
• Round 8 (next)

Results this week were fairly similar to last week’s: 6 correct guesses for “money”, 5 for each of the others. The cumulative results continue to be above 50% and inching upwards, but still nothing that can convince me this is not just a coincidence.

Here are the numbers:

I’m still not convinced you can actually make money in the long term with any of these strategies. While both “ladder” and “home” are still well into positive territory, both derive most of their gains from one single round (round 3 for ladder, round 4 for home). Of course, we still have 16 rounds to go, so there’s plenty of time for me to be convinced in either direction.

Links:

• Round 5 (previous)
• Round 7 (next)

Another fairly poor week for all strategies: 5 correct guesses for “home”, 4 for each of the others. The cumulative results still show a performance better than 50%, but just barely.

With no further ado, here are the numbers:

Links:

• Round 4 (previous)
• Round 6 (next)

A draw was the main interesting point of this week’s round. Draws are alwyas unexpected because the chance of one happening on any given game is very small; still, you can expect to see one or two on each season.

None of the strategies allows for a draw, which means that all of them missed on that game. And, overall, this was a less than stellar week, except for the “home” strategy, which got six games right and managed to get an excelent financial result. “Ladder” did 5, still a respectable result, and “money” repeated the poor performance of last week with just three correct results.

Tipping competitions deal with draws in different ways; some of them allow you to tip a draw, while other don’t; of these, some of them consider that any tip is correct for a draw, while others consider that all tips are incorrect. Since I’m using these numbers to calculate possible financial results, I’ll consider all tips to be incorrect on the event of a draw. With this in mind, here are the numbers for this week:

Links:

• Round 3 (previous)
• Round 5 (next)

In a week in which the average tipper scored 3.3 hits (according to Channel Ten), the “ladder” strategy performed amazingly well, with 6 correct guesses. In case you’re wondering, and I’m sure you aren’t, the two errors were in the Carlton-Collingwood and Bulldogs-Richmond matches.

On the other hand, the two other strategies had appalling performances, with just three correct hits each. Now, I know that the “home” strategy is pretty much random anyway (with so many teams based in Melbourne, there’s not really much of a home-turf advantage in most matches), but I was really expecting more of the “money” strategy; maybe this was a week composed mostly of upsets, after all.

On the financial front, “ladder” also did extremely well, especially because it was tipping “against the money” in most matches. The other two, though, did terribly, going well into negative numbers both for this round and in the cumulative performance (wiping the gains from the first two rounds, and then some).

Here are the numbers:

So far, I don’t think any of the strategies has proved to be good enough to bet on; the results are simply not consistent enough and, with such a small sample, any of them can be attributed to luck. Let’s see how things go as we approach September…

Links:

• Round 2 (previous)
• Round 4 (next)

Continuing the study I started on the first round, let’s see how our three strategies did this week.

The “ladder” strategy was the worst performer, with just four correct guesses (contrasting with six on the first week); no more than could be expected by blind luck. In retrospect, I believe this makes sense, and I think this poor performance may continue for a few weeks. The reason is that, after just one round (or even a few), the ladder doesn’t really say much about the relative strengths of the teams; teams will jump up and down several positions at a time until they “settle” somewhat. For the first round, the ladder represented the achievements of a whole season, which made it more representative. A better strategy would probably be a “sliding ladder”, taking into account the last X rounds (across seasons if necessary) to create an hypothetical ladder.

On the other hand, the “money” strategy did very well, with six correct guesses; the “home” strategy had five, same as last week. On the cumulative results, this puts “money” ahead on the number of correct guesses, but not on the financial results, where the “home” strategy takes the top due to its consistently good results so far (remember, no money is actually changing hands).

In short, these are the results:

More results next week.

Links:

• Round 1 (previous)
• Round 3 (next)

It’s a Melbourne thing; if you live somewhere else, you’re not really supposed to understand this. But, when March arrives, together with the new football season, comes the new football tipping season.

Everyone seems to hold tipping competitions: TV stations, phone companies, groups of friends and co-workers… everyone tries to be a football expert and to predict the results better than everyone else.

Well, I don’t really now much about Australian Rules Football, having been introduced to the game less than a year ago (and despite having read parts of “Aussie Rules for Dummies” – yes, there is such a book, and I admit to reading some sections). So, I decided to approach this tipping thing as a numeric exercise; a “technical” approach, if you will, with no more regard to the teams than what can be glanced from a few external information sources (to take the investment metaphor a little further, I am disregarding the “fundamentals” of the teams).

I devised three simple strategies; they are a starting point and can be improved for next year, if necessary. I’ll use them and follow the results during this whole year, to find out how they fare against the real “experts” out there. They are:

• the ladder: in each game, the team that is higher on the ladder (that is, has more victories and/or a better goal average) is expected to win; for the first round, the final ladder of the previous year is used; this can be rephrased as “using past results to predict future performance” (don’t try this with your investments)
• the money: as in, following it; for each game, the team most people are betting on (and, as a consequence, the one paying less to winners) is expected to win; this way, I try to use the wisdom of the masses to predict the results
• home: the simplest of all, it considers that the home team will always win; it takes home-turf advantage as the only deciding factor

In each case, I’m ignoring completely the possibility of a tie; some tipping competitions don’t even let you bet on a tie. I’m recording the results as the number of correct guesses but, also, as the amount of money I would have made if I had bet on my tips, at 5 dollars per game. I’m not actually betting, so these results are all hypothetical; maybe next year, depending on how these strategies fare this year…

After the first round, all strategies fared better than I expected; all of them did better than chance (50%) and all of them did better than the average tipper (55%, according to Channel Ten). These are the results (out of eight games):

The dollar result is the net result, subtracting the supposed expenditure ($40) from the winnings. The “money” strategy didn’t do too well, despite a reasonable number of correct tips; I believe this is a side-effect of the strategy, as it tends to bet on teams that will not pay well; as such, it needs a higher number of correct guesses to actually yield a positive net result.

I’ll post the next update in a week, after round 2.

Links:

• Round 2