Forgive me Evil Doers, it has been 29 days since my last rant. I seek forgiveness and…a hot red head. If I can only have one, then please give your forgiveness…to someone else.
Even if I receive no hot red head(s) from the above it will still have served a purpose. You see, the above illustrates that I’m horrible with introductions. The reason I’m so poor with them is that I sometimes assume everyone shares a reasonably similar foundation when it comes to various discussions. This often results in me jumping deeper into the discussion and assuming various facts are already in evidence.
Sadly, jumping ahead like that can sometimes lead to confusion. It can also have the poor result of people believing you are hiding something. In an attempt to ensure a better foundation for: ranking, MMA, red head, and math discussions in general I’ve decide to go over a few “basic” things.
Please understand that these rants aren’t the end of the road, but rather the very beginning. If I get something wrong, if I miss something, or if you just want to boast about having made it through an entire rant in one sitting, then please let me know.
As you may have guessed from the title, this rant is about rankings and how they work. Normally I wouldn’t say that, but again I’m working on not assuming too much. What follows will be a discussion of some basics of all ranking systems and some discussion of a few of the more common questions/comments I see regarding ranking particularly in MMA.
Ability Vs Achievement
One of the more common questions/statements I see regarding rankings in MMA is that people feel the lists should represent what *MIGHT* happen. Understand that the purpose of ranking systems in general is to accurately represent what *DID* happen. It’s true that accurate representation of the past can help predict future occurrences, but that’s a rant for a different day. For now it’s important to understand though that first and foremost the rankings on this site (and most other rankings) are attempting to represent what *DID* happen and not what might happen.
Now I understand there are some people that insist on bringing the “ability” or “potential” of a competitor in discussions about rankings. Let me try to provide a few examples to illustrate why that is a flawed approach.
Lets assume we have four fighters everyone agrees are ranked on “potential” as such:
1) Fighter A
2) Fighter B
3) Fighter C
4) Fighter D
Now assume that Fighter C beats Fighter B. This creates a problem because everyone agrees that Fighter B is “better”. They have more potential, they have the better skillset, etc. That means the “potential” rankings stay the same.
Since we are measuring potential it means we can’t adjust the rankings until we see something to clearly indicate that the “potential” of one fighter has dropped off or that the “potential” of another fighter has improved.
Switching sports for a second, imagine weight lifting or swimming. Sports that have absolute values in terms of performance. If one competitor sets a world record so we know their “potential” absolutely puts them at #1, but they keep losing what do we do? When using a potential driven system we’d have to continue to keep that person ranked first.
What happens in real rankings though (and on this site) is that potential only gets you so far. Ultimately it’s the performances you turn in, time and time again, that will determine your ranking.
Another great example (whether you consider it a sport or not) is golf. Tiger Woods is almost universally accepted as the “best” golfer in the world. He’s been considered the best for awhile, but the truth is that the potential exists for him to no longer be ranked #1. That won’t mean he doesn’t still have the most potential and it doesn’t mean he won’t regain his #1 ranking. What it means is that from an achievement point of view, based upon the specified criteria, he will have been surpassed (potentially however briefly) by someone else.
So the next time someone mentions, “Fighter X has to be higher because they *WOULD* totally…” do something, try to explain that what the fighter actually *HAS* done weighs far heavier than what they *MIGHT* do.
Ladder Vs Grid
Due to how ranking lists are often represented many people tend to get what I’m going to refer to as a ladder view of rankings. What we often see is:
1) Competitor A
2) Competitor B
3) Competitor C
4) Competitor D
5) Competitor E
Due to this display people often draw several faulty conclussions:
- All competitors are evenly spaced
- If a competitor improves they must overtake the person ranked higher than them
- Beating someone ranked above you must guarantee you are higher than them.
Another flaw is that finite lists have the issue of sometimes making people think those are the only people ranked, or that matter. This is a very big problem because the goal of a ranking system is often to represent the entire population involved. When talking about MMA fighters we are talking about where a fighter ranks compared to all other MMA fighters. When talking pizza we are talking about how that pizza ranks up compared to all other pizzas. Seeing a finite list of X names often makes people forget that there is an X+1 name below that, and an X+2 name below that one in many cases.
The Grid Approach
A better way to picture ranking lists is as a grid:
| C1 | C2 | C3 | C4 | Total | |
|---|---|---|---|---|---|
| C1 | N/A | 3 | 3 | 1 | 7 |
| C2 | 0 | N/A | 1 | 3 | 4 |
| C3 | 0 | 2 | N/A | 2 | 4 |
| C4 | 2 | 0 | 1 | N/A | 3 |
Often times what happens with ranking lists is we simply see the name (C1, C2, etc.) from the above and their position, but we lose all of the information in the middle. It’s that information in the middle that helps us understand the competitors overall performance compared to others and their relative position to others.
A simple example I used once before was HR leaders in baseball. Imagine if HR leaders were simply displayed as:
1) Hitter A
2) Hitter B
3) Hitter C
The above is flawed in many of the ways I mentioned before. We don’t know how close the competitors are, we don’t know if Hitter B getting another HR will move them past Hitter A, etc.
If we instead had a grid that laid out the HRs for each batter, showed us a total, etc. then we’d have a better sense of where the hitters were in relation to each other and also in regards to all hitters in general.
Odds On Favorite
I had someone mention to me not that long ago that whoever was ranked higher should be the betting favorite. This gave me pause for a minute because it showed a lack of understanding for not only how ranking systems work, but also for how betting odds work.
Since the first point covered ranking systems I’ll now take a minute to discuss betting odds. The important thing to understand though is that sportsbooks are covering their tails and adjusting to what betters do.
The first obvious sign that sportsbooks don’t care who the “favorite” is can be seen by the fact that betting lines change. Do people think this is happening because of leaked training footage of one fighter perfecting their flying gogoplata?
The reason the line is changing is because of the incoming bets. If too many people are betting on one fighter then the line will get closer. In some cases the line can even shift through and the previous “favorite” can become the “underdog”.
When things get really interesting is when a ranking system can help highlight some of these mismatched odds. That is to say situations where the potentially superior competitor is considered an “underdog”. Cases like these are fantastic for betting on because the ROI can be incredibly high.
Follow-up to point two
A bit of a corollary to the ladder vs grid point is when someone says something like:
“Fighter X *BEAT* Fighter Y HOW CANS DEY BE RANKED BELOWETH THEM!?!?oneone!!!?!?!”
Looking at the sexy grid I provided above it’s easy to see a case where one person clearly owns another yet is still ranked below them. The reason this can happen is because, as I’ve said, rankings are looking at the big picture. It’s not just about this team versus that team or this fighter versus that fighter. We are looking at overall performance across the entire sport.
Lets change sports to basketball real fast and look at the upcoming NBA Playoffs. The Cavs are the #1 seed in the east, and the Lakers are the #1 seed in the west. During the season the Lakers won their series against the Cavs 2-0. However, since Cleveland has the better overall record they have the #1 overall seed. That means they will have home court advantage throughout the playoffs. The reason for this is because the Cavs have a slightly better *OVERALL* ranking despite losing all of their games against the Lakers.
Whether either team even makes it to the finals is up in the air (pretty likely with basketball). However, the way rankings work is such that despite doing poorly head to head, Cleveland’s better performance across the board earned them a (slightly) higher rank.
Jumping right back to MMA here is another situation to consider when getting upset that one fighter is ranked lower than someone they beat. Imagine we have Fighter A ranked ahead of Fighter B after beating them. Fighter B then wins a big fight against a tough and determined Fighter C. Fighter A doesn’t fight. Fighter B then wins a second in a row. Fighter A again does nothing.
I don’t feel like typing it out, but basically copy/paste the above however many times it takes you to realize that at some point Fighter B has to be able to pass fighter A despite having lost to them head to head. Now that isn’t to say it will happen immediately and it might take numerous wins depending on how closely the fighters were ranked and who they fight. However, to immediately see someone ranked below someone they have a win over and cry foul makes you come across as someone ranked lower on common sense than a throw rug.
In a attempt to further drive this point home, lets take a quick look at the light heavyweight rankings.
1) Rashad (Even if you disagree, go with me on this it won’t matter)
2) Lyoto (Again, irrevevant, but some people have trouble if they don’t start counting at one. It’s like the people that need to do the months in order or the days of the week. *shrugs*)
3) Uhhhhh….Now what!?!?!?
See Rashad is undefeated and holds wins over a lot of good light heavyweights. Lyoto is also undefeated, so we can pencil him at #2 and not have too much conflict (other than people hating him). Once we get to #3 we are screwed in all the wrong ways.
Lets say we try to put Rampage at #3, Nope, he lost to Forrest.
Ok, lets put Forrest at #3 then…oh wait, he lost to Tito.
Fine fine, Tito at #3…oh yeah Chuck beat him….
Uhhhh Chuck to #3…DRAT~!!! CHUCK LOST TO RAMPAGE~!!!
Do you see how this goes? Without even mentioning Jardine or Wandy we can’t even make it to #3 on the ranking list without being forced to put someone ahead of someone they have lost to.
If you put Rampage at #3 (or higher) then he’ll be above Wanderlei and Forrest.
If you put Forrest at #3 (or higher) he’ll be above Tito and Jardine
etc.
etc.
At this point you are forced to limit your list to only people that haven’t lost or accept that some people will be ranked below someone they have beaten previously.