Booking Huge (In Terms of Characters) Matches

[Mongo the Destroyer]

Generally speaking, most feds will at some point run an event featuring a large match with a lot of people in them. Usually this takes the form of a rumble, though I’m sure there’s a multitude of other examples. The question then is…

How in the heck do you book a match with a ton of people?

The joke answer is: “With a bottle of alcohol and a lot of self-loathing.”

But in reality it is incredibly difficult- and still requires some self-loathing. There’s not a cure-all for any large-scale match or one-night tournament (like Call to Arms) but I can tell you what we’ve used the last couple of years for the Rumble. Because if you want to go big, you go 50+ guys writing without any limits set on them. Here comes that self-loathing again!

Since the XHF’s inception up until last year or so (as of when this was written), we used to just book the Rumble like we would any other match. Basically we’d read the billions of rps and then come together like “Well I think this guy should win” and we’d go back and forth until we got like a top 3-5 we were satisfied with and send that over to the Rumble writer.

However, it turns out that what’s good for the goose is not always good for the gander. During one of our Rumble booking sessions we (bookers: Mongo, DT, Kanyon) decided that we wanted to recognize Kuroi (the handler) for his tremendous effort with a character we hated by giving him a feature spot with a character of his we got more laughs out of- and give him 3rd place. So we did. It was uh…..not well received at all. Especially since that character only had maybe one or two promos tops and they weren’t real high-effort pieces (though not free if I recall, lol).

It seemed that another method should be employed so that we had 1. A larger final field fully booked and 2. Something more objective than “I think this guy should win.” Therefore I came up with a system that we implemented in 2020 and again now in 2021 (and likely will for the foreseeable future). The goal of this was to standardize our voting in a way that we could all agree on and would flesh out a winner (and group of runners-up) out of a pool of 50+ rpers. I think this works well for administrative groups featuring multiple bookers (who may be involved in the match) while keeping things fair and logical. Let’s take a look at the issues that I tried to address when designing the system.


Issue 1: Everyone Books Differently.

So this is a huge problem that some people might not recognize. Even among our four current main bookers (Mongo, DT, Vastrix, Dylan) you can get some wildly different opinions on how rps read. Furthermore, of the three of us who use numbered systems, how they score is all different. Vastrix for example either loves or hates an rp, so you’ll see a lot of sub 7 scores or suddenly a 9 out of 10. Now if we did everything by a ten point system, the differing weights could give one booker more power than another or accidently screw a guy who did a good job but got a 4 from Vastrix for some reason. I don’t even use numbers because I feel like any number I give is relatively arbitrary and also could be changed based on the quality of rps around it.

To address this, the new booking system would have to standardize our rating system in a way that could have weird weighting based on the numbers we happen to use that session. So instead of focusing on “who do you score the highest?” the focus became “Who are your Top Ten?” From that set we’d be able to weight people based on where they placed on each admin’s Top Ten list. Therefore, AFTER reading everything (and making notes of each and every rp via our own systems in a HUGE thread in our secret forum) each of the admins submits one of these forms:

Top 3:
1.
2.
3.
^These people will receive 3 points each

Top 6:
4.
5.
6.
^These people will receive 2 points each

Top 10:
7.
8.
9.
10.
^These guys get one point.

As noted by the form, where people rank assigns them a point value. It’s a simple system that keeps the math from getting too complex. So say for example that somebody rps SUPER well and ends up in all four booker’s top 3s. Then their score would be 3+3+3+3=12 -and they’d probably win. That hasn’t happened yet but we’ve only used this system twice so there’s always more chances in the future! One of the benefits of this system is you don’t have to specifically figure out the exact order yourself. You can be like “Oh these three guys were really awesome” and they’re all getting the same score.


Issue 2: Wait, This is Skewed Against Admins.

Yes, yes it is. For the sake of booking integrity, since the XHF’s inception (20 years ago, wow) we’ve never allowed admins to judge their own matches. Obviously in the Rumble that doesn’t work because that means either 1. Admins can’t be in the Rumble (which is problematic because we make good dross to throw out AND our rping can help encourage others) or 2. Nobody can book this match. To get around this we let admins compete AND vote BUT they can’t vote for themselves. That means that an administrator can’t be in their own top ten list.

This means that it’s nearly impossible for an admin to win. That’s sort of not motivating when again, having admins in the match is actually helpful (and honestly most of us enjoy the Rumble rp freedoms). To get around that, I added what we call a “Ghost Point.” What the ghost point does is fill in where their missing vote is. How do we decide their ghost point value? We add up their points and divide by the number of bookers who could have voted for them (the average including any zeros). Lord Dominicus’ ghost point for the 2021 Rumble was 0.3. Because of this the system is still skewed against admins, but not as strongly. So to win an admin would still need to hella impress the remaining bookers- which wow is that not easy because we all have super different opinions.


Conclusion:

This is not a perfect system (probably), but it is really good at addressing a large-sized match with a bunch of rps. Along with this we obviously add in admin comments such as who should have a good showing, some eliminations we think would work well, and sometimes we have to figure out ties (like the 2020 Rumble). I think this could also work with a single night tournament format in helping you set up brackets and figure out eliminations- though somebody would have to test that to confirm. Obviously if you are the solitary booker the only top ten that matters is your own, but this format is a way to combine/standardize several bookers’ input in a way that is both easy to set up/accomplish and gives clear, unskewed, readings.

Have a great day guys and don’t ask for your specific scores because that’s rude and probably nothing you actually want to know!

[Technical Perfection]

How we do it in the AWF.

This is not meant to counterpoint Mongo, this is just how we sort things out in our corner.

For Fired Up, our huge one night tag tournament and for the battle royal beforehand we use the following system.

Every RP gets a score out of ten. This means judges who usually mark RPs using Meltzer stars have to double their scores and judges who mark out of fifty have to fifth them. And judges who do it all on feel (ie me) have to gut up and actually put numbers down. The only thing that is important is that a judge has to be consistent with THEMSELF. For instance I picked an early roleplay last Fired Up and went "Well, that's a solid 7 in my book" and every future RP was compared to  it.

Now here's the maths bit.

We tot up every score a JUDGE has given to work out their average mark. Then we total every RP's score to get an average roleplay score for the event. Now we work out the "nudge" value for each judge, their average compared to the global average, so if the total average for every RP was 6.9 but MY average was 6.8, my nudge is +0.1. If Kuroi's average mark was 7.0, his nudge would be -0.1 Still following? Good.

Every mark given by a judge has their nudge applied to it. Adjusted scores are totalled and then you have your final rankings. It's mathematically incredibly unlikely that there are any ties in this system. It is also completely fair because Admins not judging their own matches don't get dragged down by the lack of recorded scores or up by inflated ranking positions. 

Is this system perfect? No. But it overcomes problems we've had in previous years. It also means that judges with wildly different internal scales revert to the norm. It's a very mathsy solution but once you know how to set it all up, it's MUCH easier than how I just made it sound.

Plus I love spreadsheets. I have one open right now.

[Jesse Jamester]

I personally feel if you make judging comments public, they keep people more honest - because they are then scrutinized amongst a network rather than behind closed doors.

Elaborating on this:

1. It provides everyone visibility to how judges come to their decision
2. The decision for people to 'rate' on their own merits is misdirected, because they have an audience to answer to (i.e. the people participating, not just the admins)
3. Visibility normally removes doubt of a 'Fix' being in, of someone feeling like they were cheated or didn't get a fair shake
4. It holds Admins and Judges to a standard of keeping consistent with how they judge. 


As an example to what I'm used to, for clarity sake and for a proof of concept:
community.primetimecentral.net/index.php?threads/results.41057/


Now I don't expect anything to change in the way the XHF does things. I think the system works and it works for what you have in place. But as with all things, improvements can be considered and help with cultivating a healthy community understanding of 1) how difficult judging an event of this size is and 2) what people look for in the judging process, so that the participants who want to go for it, know what to aim for. A finish line is only as good as the reason for the race. 

I appreciate everyone reading my off-the-cuff thoughts on this, so take them how you like. I'm not bitching, just sharing some thoughts. Appreciate any feedback on the topic.