Mikey Garcia - Juanma Past Statistics

[BoxStat]

I like this a lot but can you tell me how you generated/came to this conclusion?

ABILITY TO TAKE A PUNCH 63.5% (Garcia) 79.3% (Juanma)

Thanks. The figures noted are because Juanma has fought some hard hitters but Garcia still hasn't really been tested with a big puncher as yet as far as i can see. So the 'ability to take a punch' stat can be due to being untested at that level too. If we do it any other way, then it leaves it open for a fighter to fight 100 Malignaggi clones and be ranked as iron chinned, when really they've just haven't faced hard hitting opposition yet (no disrespect to Pauli, just an example ).

[AntonTheMeh]

this is actually pretty cool.

[BoxStat]

this is actually pretty cool.

Thanks. Really appreciate it, it's taken so long to make, it's nice to hear feedback that people are into it.

[.:: JSFD26 ::.]

this is actually pretty cool.

This is like some Fight Night type stats, detailed stuff.

They should add a stat where it classifies the fighter like Elite, A, B, C, D, club fighter, scrub.

(Im HALF jokingly)


Posted from Boxingscene.com App for Android

[gingerbreadman]

TS, this is a wonderful idea, but I'm slightly confused over the methods used to arrive at these numbers. Are more recent fights weighted as more important than fights several years ago, hence allowing for improvement and career trajectory? Would just be interested to see if the same method when applied to solely the boxer's last 5 or 10 fights would come up with the same numbers.

[PACnPBFsuck]

Frequent reader here and thought there might be some interest in fight stats for this weekend's Mikey Garcia bout (and others as soon as i can get them done).

It's a new website, so go easy, it's got a way to go yet. The stats compare previous bouts and their previous bouts, so is a little more in depth than other sites. It's all statistics, with no personal views or any made up rubbish:

boxstat.co/bout/68784/mikey-garcia-vs-juan-manuel-lopez

Not sure what the linking out rules are here, but can't see anything to say otherwise so i'll guess it's okay. I'm not trying to spam, so remove the post/link if it's not allowed or against any rules, and I'll be on my way . Cheeeaarrs

Let me ask you, why does it say that Alvarez data is only 59% verified, what does that mean?

[BoxStat]

TS, this is a wonderful idea, but I'm slightly confused over the methods used to arrive at these numbers. Are more recent fights weighted as more important than fights several years ago, hence allowing for improvement and career trajectory? Would just be interested to see if the same method when applied to solely the boxer's last 5 or 10 fights would come up with the same numbers.

Currently it takes into account all the fights but we'll be adding a feature to select fights from either a point in time last 1,3,5 years etc... or both fighters X previous bouts soon. With boxing it's hard to add much value to more recent fights as upsets can happen that don't necessarily mean one boxer/stat is better than another on a bigger scale. Once we get more data though we should be able to play around with the figures to see if it makes a difference and how much weight to put on more recent fights. Defo a good suggestion! For now I'll add the option to get bouts from custom dates (as it's less work) then the user can see what they need themselves if they want.

Thanks

[BoxStat]

Let me ask you, why does it say that Alvarez data is only 59% verified, what does that mean?

Hi. Because the data is used from not only the boxer and his opponents in question, but the opponents opponents too, it's an easy way for the user to see if there is any possible missing or unchecked information about them opponents too. It just saves someone trawling through each opponents information to check themselves. 50% verification is given to the boxer in question and his opponents, and 50% verification for the opponents opponents.

I'm not sure i could say opponents and more times in a reply.

[gingerbreadman]

Currently it takes into account all the fights but we'll be adding a feature to select fights from either a point in time last 1,3,5 years etc... or both fighters X previous bouts soon. With boxing it's hard to add much value to more recent fights as upsets can happen that don't necessarily mean one boxer/stat is better than another on a bigger scale. Once we get more data though we should be able to play around with the figures to see if it makes a difference and how much weight to put on more recent fights. Defo a good suggestion! For now I'll add the option to get bouts from custom dates (as it's less work) then the user can see what they need themselves if they want.

Thanks

No problem, given the amazing advance in the influence statistics have had on other sports in the last few years, it's definitely time for boxing to step it up to a level that's more than compubox. I seem to remember seeing a video detailing the GB amateur team's Olympic preparations, and it looked as if they had some absolutely amazing bits of kit with which to analyse a fighter's game, I'd try to find that video or even message them, maybe.

As a sidenote, I'd be interested in helping out in this project if that's possible, I'll pm you my email address and we could work from there?

[BoxStat]

No problem, given the amazing advance in the influence statistics have had on other sports in the last few years, it's definitely time for boxing to step it up to a level that's more than compubox. I seem to remember seeing a video detailing the GB amateur team's Olympic preparations, and it looked as if they had some absolutely amazing bits of kit with which to analyse a fighter's game, I'd try to find that video or even message them, maybe.

As a sidenote, I'd be interested in helping out in this project if that's possible, I'll pm you my email address and we could work from there?

Yeah that would be great, PM away