Re: Top 500 Players to Date
PostedCOLON Thu Dec 04, 2014 6:28 pm
Do we even know how Fantrax calculates their rankings?
The Ultimate Fantasy Hockey Experience
https://bbkl.ca/
lol, cbs had it's rankings too but they heavily favoured a high +/-. It's just a number until we see a breakdown of how it's calculated. Not to mention missing some games can hurt your ranking.kimmer wroteCOLONits pretty self explanatory bro
look at scoring in conjunction with the periphs. They regard centres in general as the most valuable, and then from there looks at the total FOW + scoring I assume with STPTS. It's not gonna be a perfect ranking system but anyone with half a brain should look at the list and deduce the rankings are based on the players ability to chip into our BBKL H2H matchups on a nightly basis with most consistency.
i think you're right. when you put your mouse over the score it says the number is based on the league's rotisserie configuration.Bruyns wroteCOLONI'm pretty sure it just weighs all categories even and assigns a point value for the ranking of that skater. For example the top +- or hits player would b e given the same amount of ranking points as the top goal or STPTS player.
It's why a player with lots of hits and PIMs and no offence can be ranked similarly to a good offensive player with no periphs. Cs are ranked a little higher than W since they get ranking points for FOW and W don't.
An email to support at Fantrax could also confirm this or explain their ranking.
So it wouldnt take into account a player like Crosby being 20 points ahead of second place in points last season. imo, fanrtax points should be given out based on a percentage of a categories sum taken by a specific player. So if crosby can get 1% of NHL points he gets 1% of fanrtax points available for that category. This would give each category equal weighting while taking into account an extreme lead like Crosby's last year. The only skater category it wouldnt work for is +/-.Bruyns wroteCOLONI'm pretty sure it just weighs all categories even and assigns a point value for the ranking of that skater. For example the top +- or hits player would b e given the same amount of ranking points as the top goal or STPTS player.
It's why a player with lots of hits and PIMs and no offence can be ranked similarly to a good offensive player with no periphs. Cs are ranked a little higher than W since they get ranking points for FOW and W don't.
An email to support at Fantrax could also confirm this or explain their ranking.
Correct, I don't think it puts any weight on how dominant they are in the category. Crosby would have only been given a small amount of ranking points more than the 2nd best offensive player.Matthew wroteCOLONSo it wouldnt take into account a player like Crosby being 20 points ahead of second place in points last season. imo, fanrtax points should be given out based on a percentage of a categories sum taken by a specific player. So if crosby can get 1% of NHL points he gets 1% of fanrtax points available for that category. This would give each category equal weighting while taking into account an extreme lead like Crosby's last year. The only skater category it wouldnt work for is +/-.Bruyns wroteCOLONI'm pretty sure it just weighs all categories even and assigns a point value for the ranking of that skater. For example the top +- or hits player would b e given the same amount of ranking points as the top goal or STPTS player.
It's why a player with lots of hits and PIMs and no offence can be ranked similarly to a good offensive player with no periphs. Cs are ranked a little higher than W since they get ranking points for FOW and W don't.
An email to support at Fantrax could also confirm this or explain their ranking.
I had my own algorithm for a few seasons, weighted to account for productivity at center/wing/d/goalie. It, like this system, was certainly subjective to an extent but it definitely helped me with valuations. Might dig it up and update it...Matthew wroteCOLONI'm sure any half thought out algorithm would have LA and Toronto's players at the top given our scoring systems. The question is how efficient is their algorithm. I am sure if we worked at it for a day or two we could come up with a far better one.