Not to complain but...

[darthpb]

Can someone explain this one?
I run a tourney of my top ranked JWW
Cervantes comes in ranked #5
He wins in first round and falls to 6th
He wins in second round and falls to 8th
He wins in third round and falls to 10th
He wins the tourney and is now rated 11th!!! And before you ask, yes, he beat good fighters...Gatti, Locche and I forget the other two but the game ratings were, I believe, two 8's a 9 and a 10.
What do you say to that?

[maxx]

If you're talking about the independent rankings, it's because the ranking system only looks at a snapshot of the last 10 fights. So, let's say your 10th fight on the list was worth 6.7 points, and in your next fight you have a good win win but the win was worth 5.4 points, you're now coming in at 1.3 points below what you were before your win. As a result, your rating may drop.

It can work the other way too. Say your 10th fight was a bad loss where you received a lot of negative points. If, in your latest fight you lose but the negative points aren't as much as the fight it's replacing, your rating can actually go up with a loss.

Hope that helps.

[guitarman]

What boxing rankings were you using? (WBA,IBF etc.) If you are refering to the generic rankings I have reported the problem several times.

[darthpb]

Yeah, the "generic" ratings...I consider it to be the "consensus" ratings, no? On the fighter screen...Here's another one...I just fought the #1 Pryor vs. #2 Arcari. Arcari beats a "partying" Pryor in the 4th(that hurt). Ratings stayed the same. Pryor is now 12-1, Arcari 19-0!!!!

[darthpb]

Maxx, which ratings are you referring to as "independent?" And I have noticed that the individual sanctioning bodies rankings do change more so that's good.

[maxx]

By independent I meant the generic ones. As a recent example of the way this works, I just had Otto Von Porat get creamed by Muhammad Ali. For the loss, Von Porat got -1.2 points. However, prior to his loss to Ali, the 10th fight on Von Porat's record was a -13.5 loss to Denver Jack Geyer. So, by losing to Ali, the -1.2 points replaced the -13.5 and Von Porat shot up in the rankings despite getting utterly dominated.

As I understand it, the rationale for this rule is to allow fighters to climb up the rankings despite a lot of losses early in their careers and, conversely, to stop fighters who won big early from staying at the top of the rankings despite a career slide later on. Under the old system, it was very difficult to overtake a fighter who had built up a lot of early wins or to bounce back from a troubled start.

As for the individual sanctioning bodies, I don't know whether they use the 10 fight rule or not.

[darthpb]

I hear what you're saying; however, it then begs the question: "WHAT?" or "WHY?"

[mking55]

I am not trying to be a rabble-rouser here.

I started a thread, as yet unreplied to the the suggestions forum a few days ago that relates to this issue.

Please check out that thread and chime in with your support.

I understand the idea behind the 10, however, I myself would like to see the number of fights used in the rankings as something that can be user-configured so everyone can run their universe the way they see fit.

[Tom Trunzo]

All good points! Rankings can be very touchy with one fight altering them in the manner previously mentioned in this thread. The 10 fight scenario is used for a number of reasons. One is that when you use all of the fights then the rankings are skewed by the most active fighter and is not realistic, another is that normally when a fighter has achieved say a top 10 ranking usually they will fight 10 fights over a course of 3 to 7 years not in 1 night like we can with the computer. So when their rankings change it does not seem so drastic in real life. Please continue with your suggestions and maybe we can throw in options such as mentioned to make it an assignable number for the user.

[Cap]

How about having the ranking show up somewhere on the individual fighter's pages?

Cap

[Speck]

What about weighting the individual "Performance Points" for each fight based on the age of the fight? The performance points for a fight (based on the fight age) would be multiplied with the "Weight" to get the modified performance points .

Example:

Most Recent Fights  Weight
1-2                 1.0
3-4                 0.9
5-6                 0.8
7-8                 0.7
9-10                0.6
11-15               0.5
16-20               0.4
21-25               0.3
26-30               0.2
31-35               0.1
36-rest             0.0

In this example, the 35 most recent fights are calculated into the boxers ranking, but an older fight has less influence on the rankings than a recent fight.

A system similar to this should control the odd effect of inverse falling/rising in the rankings.

Speck

[Kevin]

God idea Speck! That would be a much more equitable solution. The actual weighting might need to be tweaked, but it is definitely an improvement on what we have now.

[dempsonny]

I'd actually prefer a way where we could just rank them and it could be posted.

Gus

[jabbo]

Speck has the right idea. A "decay" model works best for these purposes, with older fights "Fading out" but not falling off the cliff at some arbitrary point.

Trunzos, I feel your pain. Asking the Trunzos to create the ideal fighter ranking system as part of a sim is like expecting a sci-fi writer to come up with a plausible time travel model.

Bob