Disclaimer:
-----------
First off, 'All-Time' is a misnomer.  The system cannot accurately measure
fighters from different eras against each other.  All the ratings indicate
is how a fighter compares with his contemporaries.  However, creating
all-time lists has been a long-standing hobby of fight fans over the years,
and probably always will be.  And it's fun!

(The above paragraph was extrapolated from the essay following the
"***  End  of  INTRODUCTION  ***" below.)


The System:
-----------
This System was copied pretty much 'as is' from Bob Bordier's
'BOXING REGISTRY'S RATINGS GUIDE'.  So was a lot of the description below.

A fighter enters the System and becomes Rated by beating or drawing with
an already Rated fighter. This is also called being 'Registered' in the
System.

In doing this he earns Rating points based on the quality of his opponent
and the margin of his victory.  See 'HOW TO COMPUTE RATINGS' below.

Each run of the Sytem starts with 1881 from the 'seed' fighters in
'VARIABLES'.
Those fighters (JOHN L SULLIVAN et al) are Registered from the start.

Then the system just grows, like Topsy.  Check out the early years in
'FIGHTS 1881-200n' to see how this happens.


HOW TO COMPUTE RATINGS
----------------------
The basis of the System is Computing Ratings.

                                                          WINNER   LOSER
                                                                             
1. Find LATEST RATEs from ALPHABETIC LIST and enter here  ______   ______

2. Add the two Ratings, divide by 2 and enter the results     
   in both columns.                                       ______   ______

3. In both columns enter the number of points shown below
   for the fight results:
     KO/TKO/DQ/WF within 4 rounds    160
     KO/TKO/DQ/F over 4 rounds       130
     Unanimous decision              100
     Decision                         70
     Split or majority decision       40
     Draw                              0
     Technical draw/ND/NC            not rated            ______   ______

4.  For the winner, add the entries in steps 2 and 3. 
    For the loser, subtract step 3 from step 2.           ______   ______

5. Find the number of 'RATED FIGHTS' in 'ALPHABETIC LIST'.
   a) If 8 to 16 Rated Fights, multiply Step 1 by 1.5, add
      the result to Step 4, and divide by 2.5.
      This is the fighter's new Rating.
   b) If less than 8 or more than 16 Rated Fights, add
      Step 1 to Step 4, and divide by 2.
      This is the fighter's new Rating.                   ______   ______



HOW TO COMPUTE RATINGS for non-Registered fighters.
----------------------------------------------------
To do the Rate calculations a non-Registered fighter (shown as 0000> on
the "FIGHTS" list) is assumed to have his opponent's Rate.

Therefore STEP 1 in HOW TO COMPUTE RATINGS above should read:
1. Find LATEST RATE from ALPHABETIC LIST and enter twice  ______   ______

After the calculations, if the newly Registered man's Rate exceeds
'START-MAX' in VARIABLES:
  -FOR A WIN  reduce it to the loser's starting Rate plus one.         
  -FOR A DRAW reduce it to the loser's starting Rate.



NON-HEAVYWEIGHTS SPECIAL CONSIDERATION
--------------------------------------
Starting with iteration 029 (April, 2003) I began to reduce the number of
non-heavyweights in the Site by allowing ONLY those non-heavies who
defeat or draw with Registered HEAVYWEIGHT fighters to be Registered.

Beating Registered NON-heavies will no longer get a man Registered.

The unregistered non-heavy who defeats or draws with a Registered
NON-heavy is treated specially.

He will be tagged with a Weight Category Identifier (below), and treated
as a newly Registered fighter, but FOR THIS ONE FIGHT ONLY.

If he whips another Registered NON-heavy, the same thing will happen
again.  (An example is BOBBY DOBBSw, January 1889.)

        
WEIGHT CATEGORY IDENTIFIERS:
  c  Cruiserweight
  e  Lightheavyweight (a lower case 'l' looks too much like an
                         upper-case 'I'.  See?)
  m  Middleweight
  w  Welterweight and lower.
                        

          ***  End  of  INTRODUCTION  ***

**********************************************************

**********************************************************

**********************************************************

The following ESSAY was submitted to 'THE JOURNAL' of the INTERNATIONAL BOXING RESEARCH ORGANIZATION on Feb 22, 2001, and published in issue #68.

MY WEBSITE. "All-Time Heavyweight Ratings"

For about five years I've been developing and running a computerized System that rates heavyweight fighters objectively. One of the most sinister features of boxing today is that ratings are arrived at based on anything but the merits of the fighters. Politics, favouritism, corruption all play major roles, spurred on by those excrable "sanctioning bodies" and their cretinous journalistic toadies who defile the noble game.

If OBJECTIVE Rating Systems can eventually crowd those scum out of business, then God bless us every one!

First I'll give you a little background on how the System was developed, and later on I'll explain in some detail how it works.

I started collecting newspaper write-ups of old-time fights many years ago. Of course I had to keep track of what I had and didn't have, so I worked up a file to catalogue them. That was the start of my web-site, though such a thing as a web-site didn't even exist at that time.

Jim Jacobs, the old time handball great and Mike Tyson's former mentor, insisted that the best generation of fighters was the 25 years from 1919 thru 1943. So I concentrated on press accounts from that period.

But I couldn't resist including those old-timers, from the turn of the last century, and even earlier; generations earlier than Jacobs' ideal.

Glamourous, fabulous days; Diamond Jim Brady, John L. Sullivan. The days when boxing with mufflers on the hands burst on America's consciousness with Jim Corbett thrashing the great John L. in 1892. That was a sensation!

Those were the DAYS, my friend! I couldn't leave them out!

Ten years ago I had a fairly large file of fights up to 1943. But it was just sitting there. As a subscriber to Bob Bordier's BOXING REGISTRY'S RATINGS GUIDE I began to wonder what would happen if I used his System and applied it to the fights. Would a comparison of fighters from different eras give reasonable results?

The idea began to itch as I looked over my fight records; almost 2,400 of them from 1881 thru 1943. (For that same period I'm up to near 7,000 now.)

When Mr. Bordier's publication was discontinued in 1994 I decided to scratch that itch. I'd do what Bob Bordier did, but for heavyweights only.

I'd need programs to do the dirty deed! I'm a former COBOL programmer. I programmed for a living for a while.

Well, I could get a top-of-the line COBOL compiler from California for a mere $3,000! $U.S.! Real money! Not that Canadian pretend stuff that I'm used to.

So I shopped around. My compiler, purchased in May, 1996, was MUCH cheaper.

With the programs up and running that year, I realized I'd need MANY more fights to make the results reasonable. I also was going forward into the 1950's with my fights. I started to look for help.

I brought "A PLEA FOR HELP" (published in IBRO #59, page 105) to the Boxing Convention in Canastota in 1998. Wow, wall to wall boxing fans! How could I miss? I passed the petitions out to everyone I could reach. The lack of interest was phenomenal. Not a bite. Not a nibble.

But something good happened in 1998. I met Matt Donnellon through IBRO. He seemed to be the only other man on earth who gave a damn! He was interested from the get-go, and has been a great help ever since.

I just looked at my fight file. In September of last year I started ascribing 'source' to my fights and there are 175 entries for "M.D.". That's Matt! He's been a great help with records of fighters I'd never even heard of (eg. George Byers). I think of Matt as a friend, though we've met only thru the internet. Thanks Matt.

With Matt's help the number of fight records kept growing and I kept getting more ambitious as they did. I finally brought my fights up to the year 2000.

But there was still a big gap between 1959 and the early 1980's. I have no Ring record Books between 1959 and 1983.

That's when I discovered "Heavyweights of the 60's and 70's Site" (www.geocities.com/sadmspats). A veritable orgy of data collection ensued! I was back in business.

My #3 son Timothy set up the website for me in March of last year, mainly for Matt's use. Until then I was sending him the latest run on floppy disks thru the mail.

The Site went public New Year's day, 2001, with fights from 1881 (4 fights) thru November 2000 (151 fights for the first 11 months!).

The Site's reception was gratifying and completely unexpected. Comments went from "very interesting" thru "very impressive site" and "Wow! ... very impressive" to "great work" and "extra-ordinary ... truly outstanding" and culminated in "Absolutely amazing and outstanding work... Fantastic work!"

And that just from friends and relatives! (No, no, I'm kidding!)

Here's how the System works.

A fighter enters the System and becomes Rated, or "Registered in the System", by beating or drawing with an already Rated fighter.

In doing this he earns Rating points based on the quality of his opponent and the margin of his victory. This is explained in detail in INTRODUCTION on The Site.

Each run of the Sytem starts in 1881 from the seed fighters in VARIABLES. Those seeds (JOHN L SULLIVAN et al) are Rated from the start.

Then the system just grows, like Topsy. You can check out the early years in FIGHTS 1881-2001 to see how this happens. Take your time. There's a lot of information there. First thing to do is read about the format and contents in ABOUT FIGHTS 1881-2001 on The Site.

The System calculates Rating points fight by fight and creates Year-End Ratings for each year. The fights and their Ratings are in FIGHTS 1881-2001. From these results ALL-TIME and ALPHABETIC lists are compiled along with other minor lists.

A new iteration of The Site is updated monthly and includes Ratings for the past twelve months.

The fight results are from the internet and PROFESSIONAL BOXING UPDATE. I've also found the CYBERBOXINGZONE.COM a great help, especially for old-time fight records.

The ALL-TIME list changes from month to month because not only the latest months results are updated, but also "new" fights (new to me, anyway) are being added all the time; about 500 per month lately. Some of this "new" data is 100 years old!

The All-Time list consists of the highest rating a fighter earns in the System through-out his career.

First off, 'All-Time' is a misnomer. The system cannot accurately measure fighters from different eras against each other. All the ratings indicate is how a fighter compares with his contemporaries. However, creating all-time lists has been a long-standing hobby of fight fans over the years, and probably always will be. And it's fun!

The biggest problem with the System is to keep it on an even keel. To maintain some sort of equivalence between the good old days (Corbett, Jeffries) and the good new days (Tyson, Lewis).

That's why I introduced the dreaded VARIABLES. A complex way to manipulate the data so that the results don't get too far out of line. You can check out the various VARIABLES on The Site.

My aim for now is to make the System as complete as possible (ie. add as many fights as I can find) while refining the VARIABLES to keep the results reasonable. One 'weather vane' of the System is the percentage of favourites who win over non-favourites. When that percentage starts to fall (it's in WIN % in BETTING ODDS), then I start adjusting the VARIABLES again.

Later on I intend to refine the System by skimming off from the bottom those truly inept fighters who drag the aggregate Rating numbers down. This would reduce the number of fighters rated, and of course the total number of fights.

A final word.

As I mentioned in my opening remarks, I have hoped from the outset (witness A PLEA FOR HELP), that The System would serve a serious purpose. It (or something like it) could be used as the agent for the eventual demise of those ridiculous Alphabet Soup organizations. They are more than a joke. They do positive harm to the great game by undermining its credibility. A long time ago I used to mock a friend's interest in pro wrestling by pointing out that any 'sport' that had multiple champions (there were three wresting federations at the time) could only be described as a joke. THAT WAS MY MOST DEVASTATING ARGUMENT. He was stumped. And we are stumped until we divest ourselves of those venal leeches that are sucking the life-blood from boxing.

`

Mike Paul.

michaelgpaul@hotmail.com

February, 2001