I'm trying to be realistic based on past history, you're asking that every champ be mandated to fight more often than the most active champ in history did
they could fight more often than they do but let's work up to that slowly
I'm trying to be realistic based on past history, you're asking that every champ be mandated to fight more often than the most active champ in history did
they could fight more often than they do but let's work up to that slowly
I’ll reiterate that just because they haven’t done so in the past doesn’t mean they can’t. It’s been too easy for champs to sit out because they are not being held to any standard other than Dana’s whim at the moment.
Lol, they aren't some 10-15 ranked guy who they can just match up with any of 20 opponents. There aren't 3 different #1 contenders every year for them to fight.
If they were just handing out title shots to anybody who wants one, might as well not have them at all. Say you have about 3 top 10 wins to get a title shot, then beat the champ. You think champs should have to clear out the entire top 10 every other year?
Lol, they aren't some 10-15 ranked guy who they can just match up with any of 20 opponents. There aren't 3 different #1 contenders every year for them to fight.
If they were just handing out title shots to anybody who wants one, might as well not have them at all. Say you have about 3 top 10 wins to get a title shot, then beat the champ. You think champs should have to clear out the entire top 10 every other year?
Beautifully demonstrated here by cherry picking what counts as a “year” and a “defense” to create a false perception that the fighters are less active than they are. Look, we all want to see more title fights, but let’s be honest, mmmkay?
The first fight by the champion against a challenger.
Example:
UFC 203, Stipe defended the HW title vs Overeem on 9/10/2016.
So if you defend in December and then again in February, that’s “2 title defenses in two years” due to the cherry picking misleading nature of how the stats are presented.
We are talking about Title Defense, not how many times a fighter fought for the title or reclaimed the title.
These are completely different things.
Jones winning a vacant title because he was stripped for being Jones doesn’t count as a defense even though his last 14 fights have been title fights in the same division without losing. Same for his No Contest.
I’m not confused about anything, I just pointed out the ways in which you intentionally picked the criteria in a manner which would skew the data.
You wanted to present the case that champs don’t defend enough (which I agree with as I am sure most fans do as well) but you chose a very dishonest way to go about setting your criteria to force the stats towards the result you wanted. This happens all the time, hence the quote I lead with.
You specifically chose to use numerical years knowing that it would make it seem like there were less defenses per year. It’s misleading and you know it.
Great concept for a thread, but executed in a really misleading way.
For example Stipe. You say 4 defenses in 4 years, 1 a year. But the reality is that in those years from 2016 to end of 2019 he fought 7 times in 4 years. Including 4 defenses, plus 2 title fights. That's fighting 1.75 times per year.
First of, it's not misleading. The problem is that a few people are confusing between Title Defense and Title Fight, when these are completely different.
Title Defense is when the reigning champion takes on a challenger.
UFC 203, Stipe (HW champion) defended the HW title vs Overeem (the challenger) on 9/10/2016.
Stipe gets 1 title defense.
Title Fight is when two fighters fight for the championship belt.
In the above example, both Stipe and Overeem were in the Title Fight, but only Stipe gets 1 Title Defense, because only he was the reigning champion.
If you just look at first title defense and say Stipe defends about once every 3 months, then you would be making the (inaccurate) conclusion that Stipe defends the title about 3-4 times a year...
UFC 198 Won HW title vs Werdum 5/14/2016
UFC 203 (1) Defended title vs Overeem 9/10/2016
-----------------------------------------------
First title defense in 119 days or 3 months 29 days.
This is why when calculating the average number of title defense per year, it's more accurate (and much easier) to simply look at the total years and see how many title defenses have happened.
If he happened to win the title in January there would be another defense in 2016. You would have to count the amount of months if you do it this way. Otherwise, a lot of these stats are badly skewed .
I tried your suggestion and broke it down a bit further to the days to see if it yields more accurate numbers:
Stipe
By breaking it down to the days, it showed that Stipe's career average title defenses is about 265 days / title defense, or 1.38 defenses per year.
Cormier
By breaking it down to the days, it gives the impression that Cormier, at HW, defends his belt about 1.8 times (nearly twice) per year at HW...
However, looking at his entire career title defenses (including LHW title defenses), it shows that Cormier on average defends his title every 287 days or 1.27 title defenses per year.
Conclusion: Breaking it down to the months / days still shows that the fighters defend the belt about once per year.
The one issue with looking at months/days is that it can give a false impression that a champion defends his belt more than he actually does (example: Cormier defending nearly 2 times/year at HW, when he actually does not).
This is why it's more accurate to look at the total number of title defenses for several years and take the average (the original formula from the OP).
This is why it's more accurate (and simple) to just look at the total number of title defenses for a given year and take the average (as I did in the original post).
So no matter when a champion defends his belt or loses his title, the title defense per year is still accurate and not skewed.
Example:
GSP 1st title run: (won late / lost early)
Wins title in 11 / 2006.
Defends title in 04 / 2007. -- loses title to Serra
Average title defense: 1 times per year.
GSP 2nd title run: (won early)
Wins title in 04 / 2008.
Defends title in:
08 / 2008 (Fitch)
01 / 2009 (Penn)
07 / 2009 (Alves)
03 / 2010 (Hardy)
12 / 2010 (Koscheck)
04 / 2011 (Shields)
11 / 2012 (Condit)
03 / 2013 (Diaz)
11 / 2013 (Hendricks)
-- 9 title defenses in 6 years
Average title defense: 1.5 times per year.
In some years, GSP defended his title only 1 time ('08, '11, '12). In some other years, he defended his title 2 times ('09, '10, '13).
Therefore, career average for title defense is 1.5 times per year.
Props to all champs with more than 2 title defenders that have more than 1 defence per year. Haven’t thought about it before but wouldn’t mind seeing Miocic lose his belt, I hate champs that don’t defend their belts (or aren’t active).
Based in the average number of times a champ defends his/her title per year, Miocic is on par with the average title defense per year and should not be stripped. He basically defends his belt 1 time a year, like most champs nowadays.
I don't understand why you would go to the trouble to calculate this in average number of days, and then after that, go back to the cruder, less accurate method of calculating by calendar year, where you end up counting a big chunk of the year before he wins the title and another big chunk after he loses it.
Cormier spent approximately 5 months as double champ, which obviously slows down his rate of rate of defenses somewhat since he can only defend one belt at a time.
Yes, Dillashaw really defended his title once ever 6.66 months. First defense after 3 months, second after 11 months, third after 6 months. That averages to once every 6.66 months. He held up the division once but the other two times gave a quick turnaround. The fact he fought twice more in 2016 as a non-titleholder is totally irrelevant.
First of, it's not misleading. The problem is that a few people are confusing between Title Defense and Title Fight, when these are completely different.
Title Defense is when the reigning champion takes on a challenger.
UFC 203, Stipe (HW champion) defended the HW title vs Overeem (the challenger) on 9/10/2016.
Stipe gets 1 title defense.
Title Fight is when two fighters fight for the championship belt.
In the above example, both Stipe and Overeem were in the Title Fight, but only Stipe gets 1 Title Defense, because only he was the reigning champion.
If you just look at first title defense and say Stipe defends about once every 3 months, then you would be making the (inaccurate) conclusion that Stipe defends the title about 3-4 times a year...
UFC 198 Won HW title vs Werdum 5/14/2016
UFC 203 (1) Defended title vs Overeem 9/10/2016
-----------------------------------------------
First title defense in 119 days or 3 months 29 days.
This is why when calculating the average number of title defense per year, it's more accurate (and much easier) to simply look at the total years and see how many title defenses have happened.
I tried your suggestion and broke it down a bit further to the days to see if it yields more accurate numbers:
Stipe
By breaking it down to the days, it showed that Stipe's career average title defenses is about 265 days / title defense, or 1.38 defenses per year.
Cormier
By breaking it down to the days, it gives the impression that Cormier, at HW, defends his belt about 1.8 times (nearly twice) per year at HW...
However, looking at his entire career title defenses (including LHW title defenses), it shows that Cormier on average defends his title every 287 days or 1.27 title defenses per year.
Conclusion: Breaking it down to the months / days still shows that the fighters defend the belt about once per year.
The one issue with looking at months/days is that it can give a false impression that a champion defends his belt more than he actually does (example: Cormier defending nearly 2 times/year at HW, when he actually does not).
This is why it's more accurate to look at the total number of title defenses for several years and take the average (the original formula from the OP).
Title-winning is not title defense. If that is included, it would create an extra title defense that does not exist in reality.
This is why it's more accurate (and simple) to just look at the total number of title defenses for a given year and take the average (as I did in the original post).
So no matter when a champion defends his belt or loses his title, the title defense per year is still accurate and not skewed.
Example:
GSP 1st title run: (won late / lost early)
Wins title in 11 / 2006.
Defends title in 04 / 2007. -- loses title to Serra
Average title defense: 1 times per year.
GSP 2nd title run: (won early)
Wins title in 04 / 2008.
Defends title in:
08 / 2008 (Fitch)
01 / 2009 (Penn)
07 / 2009 (Alves)
03 / 2010 (Hardy)
12 / 2010 (Koscheck)
04 / 2011 (Shields)
11 / 2012 (Condit)
03 / 2013 (Diaz)
11 / 2013 (Hendricks)
-- 9 title defenses in 6 years
Average title defense: 1.5 times per year.
In some years, GSP defended his title only 1 time ('08, '11, '12). In some other years, he defended his title 2 times ('09, '10, '13).
Therefore, career average for title defense is 1.5 times per year.
Well, if we look at his title defense frequency, it was literally 1 time per year on average:
TJ Dillashaw title run: (Bantam Weight)
Wins title in 05 / 2014.
Defends title in:
08 / 2014 (Soto)
07 / 2015 (Barao)
01 / 2016 (Cruz) -- lost title
Let's try this and see what we get:
If we break count by the months since he first won the title 05/2014) to when he lost the title (01/2016), it would be 20 months total.
3 title defense per 20 months
aka => 1 defense / 6.66 months
What I'd like to ask is, did TJ really defended his title every 6 months (aka fought 2 times per year)?
Once again, you're confusing between Title Defenses and number of fights.
TJ fought 3 times in 2016, but 2 of those fights were after he already lost his title.
Therefore, those 2 fights are not and cannot be counted as Title Defenses...
I appreciate you effort, but you are confused and flat out wrong in many ways.
I assume the purpose of this is to see how often certain fighters defend different titles. The only way to accurately count this is with month/days. It doesn't matter how many defenses someone had in 2015 or 2016 or whatever. It matters how many defenses they had from the moment they won it, to the moment they lost it. Let's take Stipe:
Won the belt on May 14, 2016, lost it on July 7, 2018. Defended 4 times in that period (Reem, JDS, Francis, DC). Total time Stipe had the belt was 784 days (as you said). And 4 defenses. So we get 1 defense every 196 days. Or, about one defense every 6.5 months. Just under 2 defenses a year. This is the reality, at least for his first HW run. You were saying he had 1 per year, which is plain wrong. This latest "run" is particular because of Corona, etc, so I'm not counting it here.
Regarding DC, again, you are right in you calculations, but wrong in your conclusion. 1.27 defenses a year is still a 27% diference. That is not marginal. 287 days (about 9.5 months) is significantly different to 365 days. This is the real stat. In other examples in your post the differences are bigger, but even here it's significant.
Your GSP example of:
GSP 1st title run: (won late / lost early)
Wins title in 11 / 2006.
Defends title in 04 / 2007. -- loses title to Serra
Average title defense: 1 times per year.
This doesn't make sense. He had a single defense. He defended 6 months after he won the title, not year. So he had 1 defense after 6 months. The prediction would say that from 11/2006 to 10/2007 he would've most likely defended twice, not once. Your model doesn't predict correctly. If he really had 1 defense a year, he woul've fought Serra on October 2007. Calendar doesn't matter, only amount of months/days.
"What I'd like to ask is, did TJ really defended his title every 6 months (aka fought 2 times per year)?"
As for your TJ example, yes, he defended ON AVERAGE about once every 6 months, or about twice a year. By "a year" I mean any period of 12 months, which is what really matters. That's the objective reality... how can you deny this when you can count the months and the defenses and divide them? I don't get it. The Joe Soto defense happened 3 months after, he took 11 months for Barao, and 6 months for Cruz... average comes out a bit over 6 months. It's not really disputable.
I don't understand why you would go to the trouble to calculate this in average number of days, and then after that, go back to the cruder, less accurate method of calculating by calendar year, where you end up counting a big chunk of the year before he wins the title and another big chunk after he loses it.
Cormier spent approximately 5 months as double champ, which obviously slows down his rate of rate of defenses somewhat since he can only defend one belt at a time.
Yes, Dillashaw really defended his title once ever 6.66 months. First defense after 3 months, second after 11 months, third after 6 months. That averages to once every 6.66 months. He held up the division once but the other two times gave a quick turnaround. The fact he fought twice more in 2016 as a non-titleholder is totally irrelevant.
That's a great question. When I did the days / months calculation as @Volador suggested, I noticed that it gave accurate averages for title defenses per year IF a fighter has a lot of title defenses (basically when there are a lot of data points to take the average). However, when a fighter has two or three title defenses, taking the average using days / months create unrealistic numbers. Please see my detailed replies to @Volador below.
In theory, it sounds like the more accurate way of calculating.
However, in reality, it can and does inflated number of title defenses per year if/when the champion defenses his title early (3 months) and then takes nearly a year off and then loses the title, such as Cormier at HW:
Note1:
The calculation using days can be accurate IF there are many data points, as in the case with Cormier's title defense at LHW.
Notice how his title defense at LHW is 1.11 / year when there are more data points to work with ?
Note 2:
When calculating using years (the formula in the OP), it gave a very similar result to the formula using days (large data set):
Years formula (the formula in the OP): 1.00 title defense / year.
Days formula with large data set: 1.27 title defense / year.
When calculating using days (small data set, fighter loses early), the result would get skewed:
Days formula with small data set: 1.80 title defense / year
Conclusion:
Using Days to calculate can be accurate IF and ONLY IF there are large number of data set. Otherwise, the most reliable and accurate for our purpose is the formula is using Years, because it works with both large and small data set and does not get skewed even if a fighter loses early.
It doesn't matter how many defenses someone had in 2015 or 2016 or whatever. It matters how many defenses they had from the moment they won it, to the moment they lost it. Let's take Stipe:
Won the belt on May 14, 2016, lost it on July 7, 2018. Defended 4 times in that period (Reem, JDS, Francis, DC). Total time Stipe had the belt was 784 days (as you said). And 4 defenses. So we get 1 defense every 196 days. Or, about one defense every 6.5 months. Just under 2 defenses a year. This is the reality, at least for his first HW run. You were saying he had 1 per year, which is plain wrong. This latest "run" is particular because of Corona, etc, so I'm not counting it here.
Stipe reclaimed the HW title on 8/17/2019.
The official travel restriction in the US came into effect on 2/1/2020.
This was the original calculation without Corona factored in: 1.38 title defense/year
This is the new calculation with Corona lockdown, as well as the reopening of the UFC (UFC 249): 1.52 title defense / year, or 1 title defense every 240 days.
Regarding DC, again, you are right in you calculations, but wrong in your conclusion. 1.27 defenses a year is still a 27% diference. That is not marginal. 287 days (about 9.5 months) is significantly different to 365 days. This is the real stat. In other examples in your post the differences are bigger, but even here it's significant.
If you are Cormier and you tell Dana / The UFC / the fans:
"I fight more often than 1 time a year. I fight 1 time every 287 days!"
It would not make any much of a difference to them, because you basically fight 1 time per calendar year. No one is going to sit down, calculate, and know that there is a 78 day difference .
Now, don't get me wrong. I'm not bashing Cormier. What I'm basically saying is that if a fighter defends the belt 1 time per calendar year, to the public, it doesn't matter if it took 287 days or 365 days.
Defends title in 04 / 2007. -- loses title to Serra
Average title defense: 1 times per year.
This doesn't make sense. He had a single defense. He defended 6 months after he won the title, not year. So he had 1 defense after 6 months. The prediction would say that from 11/2006 to 10/2007 he would've most likely defended twice, not once. Your model doesn't predict correctly. If he really had 1 defense a year, he woul've fought Serra on October 2007. Calendar doesn't matter, only amount of months/days.
Basically, there are 2 ways to calculate the average number of title defenses per year:
1) By days / months
2) By calendar year
Calculating by days / months are accurate IF there are many data points to work with.
However, if there are too little data points, such as when a fighter defends early within a few months and loses, the result would be inflated.
For example, when GSP defended his title in 140 days (4 months 20 days) and lost to Matt Serra, doing the calculation by days / months would have you believe that GSP defends his title 2.61 times per year:
This is not accurate, and I will prove it below!
With many more data points (GSP's entire career), the result shows that GSP defends his belt only 1.68 times per year:
...
Now, if you remember, previously, I used the calendar year for the Matt Serra example and got 1title defense / year.
1 is closer to 1.68, than 2.61!
My point? Using the day / month in the calculation does NOT work when there are not enough data, such as when a fighter defends a few fights!
Therefore, when there are only a few data points, it is more accurate to use the Calendar year, NOT the day / month formula!
As for your TJ example, yes, he defended ON AVERAGE about once every 6 months, or about twice a year. By "a year" I mean any period of 12 months, which is what really matters. That's the objective reality...
how can you deny this when you can count the months and the defenses and divide them? I don't get it. The Joe Soto defense happened 3 months after, he took 11 months for Barao, and 6 months for Cruz... average comes out a bit over 6 months. It's not really disputable.
When doing average calculation using days and there are a small number of data sets (3, as in the case with TJ's title defenses), it incorrectly inflates the number of title defenses that a fighter supposed to do.
When Jones got suspended and lost the belt, the time before the suspension is counted, but the time from the suspension onward is not counted against him.
Here is the formula using days / months as @Volador and a few others have suggested: Jones was truly one of those rare champs who defended his title exactly 2 times per year on a consistent basis:
As you say rounding by years is rough and can be misleading
Johnson was champ for 2142 days and defended 12 times, that includes the unsuccessful defense against Cejudo, average time betwen defences equals 178 days. Rousey was champ for 1074 days with 7 defences giving 153 days. OK she was gifted the belt so look at Joanna 966 days with 6 defences gives 161 days.
Of course Mighty Mouse maintaining that 178 day average over more than 2000 days is the greater achievement even if Rousey takes the lead, like they say lies, damned lies and statistics.
Luckily wikipedia's list of UFC champs has already done the hard work of detailing lengths of reigns in number of days.
As for champs who only defended once or twice, just exclude them, we don't want Esparza at 92 days topping the list.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
.
