OVAH NERV Ranking

[KCU Master 2007]

In a push to get the NERV scoring system to a more national level, OVAH has jumped on board. This thread will list the scores for OVAH area members only (if your not a member of OVAH but still compete in the events, you'll be NERVed).

The current rankings are only from the past Dayton Tournament on May 16th, but there are a few others in the mix.

Here are the current rankings for OVAH NERVs:

KCU Master 2007**              1291.86
Clarissimus*                         1262.07
Phantazm21zero**               1259.34
Outforblood**                       1245.51
Allison*                                 1236.41
Wookie*	                          1236.14
MattserTruckRally*             1236
Andrew*                               1235.59
Will*                                     1234.86
Father of Julian**                1221.93
Odin Osgard*                       1214.07
Steve Ullum*                        1213.66
killercactus**                      1212.4
Aldin*                                  1212
Brownsfan82*                      1212
Jacob_j_p*                          1212
Jorundr Shieldcrusher*       1212
Mechabeast*                      1212
SuperflyTnT*                       1212
The Whisper**                     1212
Darkness1717**                  1201.24
southwest ninja***              1191.52
Marduk*                             1189.66
Richard*                             1188.41
Drea*                                  1188
Heath*                                1188
Moxilla*                              1188
Noxious Ghoul*                  1187.59
Steve Mayne*                     1187.59
Jon Duke*	                        1187.17
Randy Pitstick*	               1186.17
Sarah**                               1183.95
Q9 (Julian)**	               1178.9
Jacob*	                        1176
Raudulfr Shieldcrusher*    1176
Cory*                                  1167.45
Derek*                                1164.69
Jay                                     1164
Luckyfingers09*                 1164
OH-MEGA*                         1164
Opal*	                                 1164
Ian*                                    1160.55
Steve**                               1159
Jay’s G-Pa**                      1153.24
Tony*                                  1152
TJ**                                    1143.73
Hannah*                             1139.17
Brad**                                1125

This list will grow as tournaments continue to take place.

To look at the NERVs of scapers in other regions check the following links:
North East Region
National Rankings

FAQ

What are these numbers?
They are Elo ratings. Briefly: as you win games your rating goes up, as you lose games your rating goes down. However, the size of the adjustment takes into account the strength (current NERV) of your opponent---you get a bigger boost for taking down someone with a high rating.

How do the regional lists and national lists interact?
The same algorithm is used in each region. If there are not many games between different regions then it makes very little sense to directly compare the NERV of a player in the first region with one in the second. However, as more inter-region games are played---at GenCon for example---we hope that at least some meaning might be extracted.

How is it calculated?
Every new player starts with a rating of 1200. If Player A has a rating of X and Player B has a rating of Y, and Player A beats Player B in a tournament game, then Player A's new rating is given by X + K(1 - E), where K is a fixed parameter (I am using K=24), and E is the expected result of A, given by (1/(1+10^((Y-X)/400)))). The calculation is similar for B, but with a 0 in place of a 1 in that first formula.

Ick! I don't like math. Should I stop reading?
No. You don't need to understand the nuts and bolts to understand the big picture. If two players with equal ratings face off, then the winner's rating will increase by 12 and the loser's will decrease by 12. If they tie, their ratings will remain the same. If Player A has a NERV that is higher than Player B's, then the change in rating will be different. The winner's will still go up and the loser's will go down by the same increment. However, the increment will be larger than 12 if B wins and smaller than 12 if A wins. The increment is never larger than 24.

If you play a lot against someone with a (stable) NERV 100pts lower than your (stable) NERV, you should expect to win a shade under two-thirds of the time. Against someone rated 200pts lower, expect to win three-quarters of the time. This reasoning can be reversed: If you have a regular opponent against whom you win 2/3 of the time, then you should expect your NERV to settle around 100pts higher than his/hers.

Mmmmm numbers. Do you have any more?
PM me with your email address and I'll send you the latest version of the spreadsheet that has all of the data and calculations in it.

How do I get on your NERVs?
Simple! Attend a NERV-approved tournament. Attending an "open event" (see below) is the only way to initiate your NERV adventure.

Do these numbers tell us how good someone is at Heroscape?
Not really. It will take many more games to settle to a "true" level. In any case, there is plenty of noise (choosing 'fun armies' over 'competitive armies', variations in tournament style,...) in the system that might swamp the "true Heroscape skill level" that we are trying to measure.

In an odd quirk, it turns out that these numbers are an accurate reflection of ones worth as a human being. Weird, huh?

What counts as a game for the purposes of this list?
There are three classes of game:

1. Open Events. If it's a one-on-one Heroscape tournament and has at least three attendees that already have a NERV, then I'll probably include it. It should be event advertised in the Events section of heroscapers.com to which all are welcome (subject to space and other constraints, obviously). The only reason not to include one would be that it has conditions that are too unusual, but the variations we've seen recently (army size variations, glyphs or no glyphs, Marvel or no Marvel, fractional or whole card scoring, sidebar) are all well within the tolerance. Any decision on whether a tournament is included will be made arbitrarily by me. My decision is final, unless I change my mind.

2. Private Games suspended until further notice. Private Games. In addition to games at open events, each NERVer may play up to four private NERV-rated games per month against other NERvers. No opponent may be played more than twice in a month. Both players must agree prior to the game that it is to be NERV-rated. The game must be one-on-one kill 'em all Heroscape with armies in the 400-600pt range and no special rules. It is up to you whether you want to use glyphs, Marvel, drafting or premade armies, a time or round limit or starting zone limitations, as long as both players are fully aware of the parameters. The result of the games must be reported in this thread. [This category of game was introduced in August 2008; depending on how it goes the number and prameters of allowable games may be changed.]

3. Other. I don't expect this category to be used much; it's here so that I have the wiggle room to make decisions on types of event that have not yet been considered without people thinking I'm breaking the rules. The only known acceptable game type here is a GenCon tournament game between two players with existing NERVs. Established leagues and open game days may also feature and will be considered on a case-by-case basis as they arise.

My region isn't represented. Do you hate me?
If you want your region to be represented, you will need to have someone step-up to maintain the NERVs for your area. We are actively looking to recruit new regions. PM me if you are interested in taking the next steps in your region.

Why are you doing this?
I was inspired by CaptainCanuck's national ranking thread. I'm far from convinced that either that exercise or this will lead to any insight into Heroscape ability. However, I do think that there is at least some information to be gained---and fun to be had---in tracking both systems.

Aren't there going to be lots of players that only attend one or two tournaments and so their level will hover around 1200, cluttering up the table without it being a true representation of their skill?
I will clean the list up periodically. All players with no activity for more than 1 year will be rendered "inactive". This will remove them from the list and mean they cannot play in closed events. If an inactive player plays in an open event, the stats will be resurrected and the player carries on as a NERVer from where he/she left off.

If someone plays twice as many games as me, won't they have an advantage?
No. After an initial period of heading towards your approximate rating level, you will stay more-or-less steady. For example, if Player A has settled NERV of around 1300 and Player B has a settled NERV of around 1100, then Player A will win about 3/4 of the games. If Player A wins, then Player A gets a boost of 6 and Player B gets docked 6 points. If Player B wins, then Player A gets docked 18 points and Player B gets a boost of 18. In other words, the ratio wins:losses is 3:1 for Player A, but the ratio of increments is 1:3. These cancel each other out, and both players maintain their approximate rating level.

The way to improve your rating beyond that point is to improve your play. If your skill level increases, your NERV will too. I see this as the primary use for the NERVs: tracking your progress (improvement, one would hope) over time. To do this, don't rely on the fluctuating principal list; track the list of established ratings.

Why "New England" if this is a national system?
This reflects the historical roots in the New England area. We play Swiss tournaments outside Switzerland; now we have NERVs across the country.

Who is behind all this?
I (KCU Master) maintain the OVAH Region. Ollie maintains the North East region list and compiles the lists from different regions. Other shadowy figures help out behind the scenes.

Can I go now?
Yes, thanks for reading.