Given the early release of their progressive findings, I thought i'd take the opportunity to give you all a brief insight into my own personal 'prediction model' and how it differs to that used by Champion Data. I'm sure a lot of you may have differing opinions to me, but after all, in this caper, it's all about tapping into the umpires minds and attempting to think like an umpire.
Based on what they have coined a 'prediction model', their formula is both founded upon and heavily weighted towards Supercoach rankings, among a range of other varying factors. Of these others, includes the theory that umpires are more likely to favour run-and-carry type midfielders who kick long.
In comparison to Champion Data, my ever evolving model for distribruting votes is largely founded upon the theory that in-and-under/contested possession/clearance type midfielders are the sort of players that are most likely to first catch the eye of the umpiring fraternity ahead of those run-and-carry players I would otherwise term 'possession accumulators' or 'link-men'.
The classic example of this in effect is Swan's polling performance at the Brownlow last year. Now I am a fan of Swan as a player, however, his style of play is such that I don't think it does him any favours when it comes to featuring in many of the umpires votes. You'll find he is one of these 'possesion accumulators' that I talk about, who stand on the outside of stoppage packs and are used as the 'link-men' in a passage of play once the in-and-under teammate obtains possession of the ball and dishes off to them.
So I am of the belief that whilst these players may still have an influence on the game, they do so indirectly as opposed to in-and-under players who dive into the contest and directly cause a turnover/retrieve the ball from a stoppage and give their team first use of the ball in the form of a clearance. Thus, the first player to catch the umpires eye at a stoppage is the in-and-under/contested possession/clearance specialist BEFORE the run-and-carry player who is used as a link-man in driving the ball forward.
Many were convinced last year that by gathering 35+ possessions each week Swan would therefore have performed considerably well at the Brownlow. Early on for a short period last season I too thought Swan would've come close to taking home the Charlie - that is until I started to monitor his game more closely. The truth is, that while he is prone to winning his own ball at times, he gets a majority of his possessions in his team's defensive half; many of which are uncontested. I think it's this aspect of his game that may perhaps look unappealing to the umpires and - whether rightly or wrongly - translates into 'cheap possession football'.
Of the other factors that Champion Data uses in formulating their predictions - and one which I am in agreeance with them on - is the contention that the umpires are, on the whole, more favourable towards midfielers over forwards/rucks/defenders. The Brownlow is still a midfielders medal, and I don't think there are any signs indicating this will change this season.
Now Champion Data have had a somewhat decent track record, and they always provide an interesting perspective on the Brownlow, but I wouldn't say that they have been impeccable in predicting past winners. Sure, they correctly chose Bartel in 2007 and Ablett in 2009, but they failed to select Cooney outright in 2008, instead hypothesising that he'd finished 3rd. You'll notice that both Bartel and Ablett were both in the top two favourites to take out the Brownlow for their years, so CD's correct predictions aren't all that surprising, whereas Cooney was largely seen as a shock winner in 2008 - which they failed to predict like most of us.
One other glaring discrepency in their formula is the fact that they don't take into account the scoreline for each match. So, you may find that they give 3 votes to a player whose side gets smashed by 10+ goals, and wasn't otherwise as deserving of the votes ahead of a player from the opposing side who won.Hence, I wouldn't read too much into their findings, but instead urge people to use it merely as a very loose guide.
Nevertheless, I look forward to their final results come the end of the season and how it compares to mine.
If you have any theories or formulas that you adopt yourself, i'd love to hear them. Feel free to post them below.
NOTE: My computer harddrive that my phantom tally is saved on was infected with a virus last week, and thus, at the moment it lays on there dorment until I get around to transferring it across to my new harddrive. So at the moment i'll be updating the leaderboard that was posted in the previous round, and consequently it may not be as accurate given I cannot remember who lies a vote or two outside of the leaderboard I posted, which means that players may now have worked their way onto the leaderboard that I'm unaware of. I'll update these players once I retrieve my word file containing the full leaderboard. Cheers!
Port Adelaide V Collingwood:
3. D. Rodan
2. D. Swan
1. S. Pendlebury
Geelong V Hawthorn:
3. J. Bartel
2. J. Selwood
1. C. Rioli
West Coast V Adelaide:
3. S. Thompson
2. S. Goodwin
1. M. LeCras
Brisbane V St. Kilda:
3. L. Montagna
2. L. Hayes
1. S. Black
Richmond V Fremantle:
3. D. Connors
2. D. Martin
1. C. Newman
Sydney V North Melbourne:
3. A. Goodes
2. N. Malceski
1. T. Kennelly
Melbourne V Essendon:
3. B. Green
2. C. Sylvia
1. M. Jamar
Carlton V Western Bulldogs:
3. M. Boyd
2. B. Lake
1. A. Cooney
-----------------------------------------------------------
Round 15 Leaderboard:
18 - L. Hodge
16 - C. Judd
16 - J. Bartel (+)
15 - L. Hayes (+)
15 - G. Ablett (-)
15 - M. Barlow (-)
14 - D. Swan (+)
12 - S. Pendlebury (+)
12 - A. Cooney (+)
12 - L. Montagna (+)
12 - M. Boyd
11 – P. Chapman
11 - A. Sandilands
11 - N. Dal Santo
11 - J. Watson
11 - B. Goddard
10 - B. Harvey
9 - J. Selwood (+)
9 - J. Bolton
9 - A. Swallow
Note: (*) Ineligible. (+) Ranking Rise. (-) Ranking Drop.
'07 Bartel was not in the top 2 fav's. He was paying $10 that yr. How do I know, because i had $100 on him
ReplyDelete