User Pass
Home Sign Up Contact Log In
pulse
offline
Link
 
Because everyone loves a friendly neighborhood math geek!
(also, hooray for excel)

RPI calculated using 25% winning perc, 50% opp. win percentage, and 25% opponents' opponents win percentage. it's pretty standard.

sorry for the formatting... these forums aren't too friendly to tables.

Here we go:
# C RPI (W-L-T) Team
01 E 59.64 (12-1-0) Sharks
02 W 58.78 (12-1-0) Monkeys
03 E 58.28 (13-0-0) Tigers
04 W 58.13 (13-0-0) Pewnts
05 E 57.54 (11-2-0) Sin
06 W 56.17 (11-2-0) Wildcats
07 W 55.91 (11-2-0) H. Bayou Bengals
08 W 55.41 (9-3-1) Bolts
09 E 55.31 (11-2-0) Grizzlies
10 E 55.23 (8-5-0) Wolves
11 E 54.34 (10-3-0) Taints
12 E 52.42 (7-6-0) Crabs
13 E 51.95 (8-5-0) Pythons
14 W 51.84 (9-4-0) Raiders
15 W 51.00 (7-6-0) Bigga Bois
16 W 50.28 (7-6-0) B.A. Marauders
17 E 49.23 (6-7-0) Condom Raptors
18 W 48.49 (6-7-0) Hangovers
19 W 47.39 (5-7-1) LUEshis
20 E 47.05 (4-9-0) Farvas
21 W 46.81 (4-9-0) Crusaders
22 E 46.58 (4-9-0) Kings
23 W 45.96 (3-10-0) Storm
24 E 45.84 (3-10-0) MA Marauders
25 E 44.62 (3-10-0) Virus
26 W 44.86 (3-10-0) Cougars
27 E 44.19 (3-10-0) B.R. Bayou Bengals
28 W 42.92 (2-11-0) Anarchs
29 W 41.60 (1-12-0) Lesbians
30 E 40.10 (1-12-0) Buckeyes
31 E 37.59 (0-13-0) Bayhawks
32 W 36.68 (0-13-0) Knights

Anyone else shocked by the #1?
Last edited May 22, 2008 00:03:40
 
trojans4me
offline
Link
 
No, Sharks rule!!! <chomp chomp>
Nice work on the math by the way.
 
Link
 
I would be shocked except the only good team they haven't played yet is the Taints, whereas we haven't played the Sin or Grizz. If we beat the Sin tomorrow, that should move us above the Sharks...I would think.

I'm guessing it's the same with the Monkeys and Pewnts because I haven't looked.
 
Iktomi
offline
Link
 
Nice work. No surprise here. It would be nice for you to do it again after week 16 when we have the cross-confrence match ups.
 
PIG~
offline
Link
 
:/
Last edited May 22, 2008 07:42:28
 
pulse
offline
Link
 
i think the shocking part for me wasn't so much that they're on top, but rather the margin... they're a full point ahead, and that's not a small difference in such a tight group (the next 4 teams are all within 1 point of each other).

even with a win over the sin, i'm not sure the tigers will go that high, though i expect the monkeys won't be #2 for much longer.

anyways.. i'll update this week to week, it's pretty easy now that the excel sheet is made. I'm gonna have to find a different way to post them though, 'cause that was a pain in my butt.
Last edited May 22, 2008 08:30:32
 
coachviking
offline
Link
 
Now pulse I'm guessing that you think the EVIL MONKEYS won't be two for long because they will be moving up to #1, correct? LOL
 
Link
 
The math is bad, it should be 50% win pct, and the rest 25%. Remember winning is what counts.
 
pulse
offline
Link
 
Originally posted by Roll Fizzlebeef
The math is bad, it should be 50% win pct, and the rest 25%. Remember winning is what counts.


no.... what counts is winning against teams that win. otherwise it would simply mirror the standard rankings based on records.

the math that i didn't include (because i suck at excel) is that when calculating the OWP, you're supposed to ignore games played against the team. For example, when calculating the RPI for the pewnts, i need to use the WP of the evil monkeys (averaged with every other team they played). Since the monkey's only loss was to the Pewnts, and we are ignoring games played between the two (for this calculation), then the monkeys WP would be 1.000. Every team the pewnts played they beat, lowering that team's WP. The average of all those WP's makes up the pewnts OWP, so it is lower than it should be. so basically... these rpi's are not entirely accurate.

That takes a lot of friggin work to implement, especially this late in the season, and my guess is it would require a separate spreadsheet for each team. We'll see if i get lazy enough to actually spend that much time fixing it... or if i just keep running with the (broken) system.
Last edited May 22, 2008 10:51:07
 
Link
 
Originally posted by pulse
Originally posted by Roll Fizzlebeef

The math is bad, it should be 50% win pct, and the rest 25%. Remember winning is what counts.


no.... what counts is winning against teams that win. otherwise it would simply mirror the standard rankings based on records.

the math that i didn't include (because i suck at excel) is that when calculating the OWP, you're supposed to ignore games played against the team. For example, when calculating the RPI for the pewnts, i need to use the WP of the evil monkeys (averaged with every other team they played). Since the monkey's only loss was to the Pewnts, and we are ignoring games played between the two (for this calculation), then the monkeys WP would be 1.000. Every team the pewnts played they beat, lowering that team's WP. The average of all those WP's makes up the pewnts OWP, so it is lower than it should be. so basically... these rpi's are not entirely accurate.

That takes a lot of friggin work to implement, especially this late in the season, and my guess is it would require a separate spreadsheet for each team. We'll see if i get lazy enough to actually spend that much time fixing it... or if i just keep running with the (broken) system.


Monkeys lost to the Bengals, not the Pewnts.

C'mon, we may be in the East but you gotta pay attention...just playing.
 
pulse
offline
Link
 
ahahahaha. well i'm not rewriting that example... so just pretend.
 


You are not logged in. Please log in if you want to post a reply.