Manny Pacquiao is a More accurate Power puncher than Floyd Mayweather.

[crazy8s]

People you have to be careful with statistics.

Numbers don't lie but they can be manipulated if not presented properly. I'll give you an example:

I saw a video of a guy comparing Kobe and LeBron's career stats. One stat showed that Kobe had exactly twice as many games shooting under a certain percentage (I think it was under 40%) as LeBron had.

The thing he failed to mention was that, at the time, Kobe had been playing for exactly twice as many seasons as Lebron had. Leaving out information like that is manipulative.

My point is, you be need to be more careful with statistics. If you don't include the right information (or in this case don't even compute the numbers correctly), then your stats are worthless.

[Higher Animal]

Why is compubox inaccurate? I've never seen or heard it discussed in a logical manner.

Appealing to the majority (Fuh Fuh Floyd is awesome cuz people say so) is stupid, in my opinion.

It seems valid to say Mayweather is the more accurate puncher overall with better defense while Pacquiao lands a slightly higher percentage of harder shots (only somebody who is a cunt will argue that a jab is equivalent to a hook or cross) that probably hurt more due to Pacquiao having more power.

I'll go ahead and say it: if TKOs and KOs are considered victory then the difficulty and hardness of a shot has to be considered in the formulation of points scored. In another context, Floyd manipulating his opponents by leading them and dodging them (IE ring generalship) should also score regardless of hits per punch or total volume landed. If you have a simple calculation of overall hits landed then you'd have to award Pacquiao the undisputed victory over Marquez last year.

Boxing in ancient Greece consisted of two people fighting to the death or quit. In a more humane century the sport has to consider the multiplicity of inputs into scoring that come with a modern appreciation of the sport.

[nipplefloss]

I don't understand why you average the averages to establish this, that's a really goofy way to define average accuracy.

If I fight two fights and I'm 1/1 in the first and 500/1000 in the second then my "average" accuracy by this method is 75% but my actual average accuracy if you average over all punches thrown is 50.004%. No one in sports does averages this way because it is incredibly stupid and would lead to really stupid results based on outlier performances.

This ignores the fact that compubox itself is just a dude watching a fight and pushing a button to determine when he thinks a punch landed, so saying "you shouldn't trust your eyes, you should trust the numbers" is silly because the numbers are based on some guy using his eyes, as well or as badly as the rest of us. Basically your method of establishing your premise is flawed and your source for numbers is also flawed. And why stop at only 8 fights? I'm willing to bet that you can get the numbers to say a lot of different things depending on when you choose to set your cutoff for fights that qualify.

[Cormega]

Originally Posted by nipplefloss

I don't understand why you average the averages to establish this, that's a really goofy way to define average accuracy.

If I fight two fights and I'm 1/1 in the first and 500/1000 in the second then my "average" accuracy by this method is 75% but my actual average accuracy if you average over all punches thrown is 50.004%. No one in sports does averages this way because it is incredibly stupid and would lead to really stupid results based on outlier performances.

This ignores the fact that compubox itself is just a dude watching a fight and pushing a button to determine when he thinks a punch landed, so saying "you shouldn't trust your eyes, you should trust the numbers" is silly because the numbers are based on some guy using his eyes, as well or as badly as the rest of us. Basically your method of establishing your premise is flawed and your source for numbers is also flawed. And why stop at only 8 fights? I'm willing to bet that you can get the numbers to say a lot of different things depending on when you choose to set your cutoff for fights that qualify.

[ame]***********.youtube.com/watch?v=Y2aTd-coddI[/ame]

Here Pacquiao lands 19 punches in a round and compubox gives him credit for landing 35. Anybody could have done a better job than that sitting at home watching the fight on TV and counting punches landed on their fingers.

[D4thincarnation]

Anyone willing to provide me with proof on the inaccuracy of compubox?

I've counted rounds myself and found compubox to be extremely accurate.

[Cormega]

Originally Posted by D4thincarnation

Anyone willing to provide me with proof on the inaccuracy of compubox?

I've counted rounds myself and found compubox to be extremely accurate.

Check the post directly above yours. If that isn't convincing enough, go on youtube and you'll find countless rounds with compubox stats shown at the end. Count the punches yourself and you'll find that they're often times way off. Pacquiao-Bradley round 10 I believe is another example of one guy (Pacquiao) getting credit for landing nearly double what he actually landed.

[D4thincarnation]

Originally Posted by Cormega

Check the post directly above yours. If that isn't convincing enough, go on youtube and you'll find countless rounds with compubox stats shown at the end. Count the punches yourself and you'll find that they're often times way off. Pacquiao-Bradley round 10 I believe is another example of one guy (Pacquiao) getting credit for landing nearly double what he actually landed.

I have counted rounds myself and found compubox to be very accurate.

Can you provide me with evidence where compubox is not accurate.

[Cormega]

Originally Posted by D4thincarnation

I have counted rounds myself and found compubox to be very accurate.

Can you provide me with evidence where compubox is not accurate.

I just gave you two rounds. One I posted a video of. If you bother to watch it and you still find them to be "very accurate" then you're just a complete ****ing moron and that can't be helped.

[D4thincarnation]

Originally Posted by Cormega

[url]***********.youtube.com/watch?v=Y2aTd-coddI[/url]

Here Pacquiao lands 19 punches in a round and compubox gives him credit for landing 35. Anybody could have done a better job than that sitting at home watching the fight on TV and counting punches landed on their fingers.

Sorry is there a link on this post.

Couldn't see it, does not show up on my phone.

[carrotlad]

Scientific statistics have no place in boxing.

These statistics are heavily flawed.
There is a case to be made that it is harder to land shots as shots thrown increases (Pacquiao)
but that's about as far as the comparison can be truly drawn.

If we look at these 2 boxers and compare them as separate and complete subjects (each defined by having their
specific skills, styles and dimensions) then many of these variables can be eliminated.
[What this means is that you cannot draw upon one characteristic of either boxer and determine how it affected the results]

If they are compared as a complete article, styles and size of subject can be overlooked.
Variables to be controlled depend on:
1) The people the subjects have fought would have to be same. In a true test this would be the independent
variable; what we would change in order to see it's effect on the dependent variable (accuracy).

2) How the opponents fight the subjects. Perhaps a less obvious one, the style in which the opponent fights
the subject will greatly affect the accuracy. This would have to be controlled as best as possible. It's very common for
a boxer to fight one opponent one way, and a different opponent another way. To be a completely fair test
the opponent would have had to have fought the exact same fight for both subjects.

3) Physical Factors (size of the ring, glove size, weight class). Need to be controlled in order to to
ensure they do not affect the results. They would have to be the same for all fights compared.

4)Other Factors (referee, crowd, location). All may affect the mental condition and therefore performance
of either the subject or opponent.

Hopefully you will have seen there is no way a true scientific level of testing will ever be done. Statistics have no place in boxing, outside the financial aspect.

[Cormega]

Originally Posted by D4thincarnation

Sorry is there a link on this post.

Couldn't see it, does not show up on my phone.

There's a youtube vid in that post... Pacquiao-Bradley round 4 complete with a counter for Pacquiao's landed punches. He CLEARLY only lands 19 and compubox gives him credit for landing 35.

[Hands of Iron]

Originally Posted by D4thincarnation

Sorry is there a link on this post.

Couldn't see it, does not show up on my phone.

Your phone sucks.

[D4thincarnation]

Watched the vid and the guy misses a load of punches that land, I got 36.

When you throw a punch and the glove hits a scoring zone body or head, it lands.

[Higher Animal]

Originally Posted by Cormega

Here Pacquiao lands 19 punches in a round and compubox gives him credit for landing 35. Anybody could have done a better job than that sitting at home watching the fight on TV and counting punches landed on their fingers.

It's round by round as it happens. You'd have to make the argument that an unofficial hit keeper is using his influence to bias events. To dismiss out of hand is to simply dismiss all judging that doesn't comport to your personal bias.

I'm willing to accept compubox as inaccurate; but it's difficult to make the argument that it's inherently biased and subjective like boxing judging. Even the example shown slowed down I saw some missed potential punches on the part of the youtube reviewer. Not that I'd call them punches necessarily but just that it's possible for that to have been the case

[D4thincarnation]

Originally Posted by Cormega

There's a youtube vid in that post... Pacquiao-Bradley round 4 complete with a counter for Pacquiao's landed punches. He CLEARLY only lands 19 and compubox gives him credit for landing 35.

I got 36 from that clip, got the full version on hi res, I'll check it again but expect to see a number either side of 35 or 35 itself.