Photo matching from pinstripes

[Jason]

I come from the NHL jersey collecting scene, so forgive me for my being a newbie on this question.

A big benefit of collecting hockey jerseys is that they are fairly easy to photo match due to the many markings they can accumulate during games (stick marks, puck marks, tears, holes, fabric pulls, team repairs, paint transfers etc).

I have seen quite a few threads showing supposed photo matches showing certain pinstripes as the identifying marks only. Being a newbie to baseball jerseys, my guess is that logo's and patches may be sewn on slightly differently on each jersey so the pinstripes may appear in different locations from jersey to jersey. Is this correct?

If the above is correct then my next question would be....is this really a reliable way to photo match? Lets say a player wears 10 jerseys, all the same size and specifications, is it safe to say that 2 (maybe even more) of those jerseys may have pinstripes in identical positions using logo's and patches as references?

Unless I am missing something, using pinstripes would not be justification for calling a jersey "photo matched". I know we all might see things that aren't there in the photo because we want that photo match so badly..but my opinion is that if it is not a unique, distinct marking in the photo then it is not a 100% photo match.

[hiramman]

I'm by no means an expert on photo matching, however I do know that on recent Yankee home pinstripe knits that there are at least a couple of different specific areas to check when trying to photo match a Yankee pinstripe.

1)placement of Yankee logo in reference to the pinstripes.

2)placement of button holes in reference to the pinstripes.

3)placement of pinstripes showing around the collar from the inside neck reinforcement material. (On the inside of Yankee jerseys the second layer of material sewn around the inside of the neck/collar area is a seperate piece of pinstripe material. The ends of the pinstripes on that material will show up as black marks along the outer collar seam because the inner pinstripes will not line up with the outer jersey pinstripes.)

4)placement of the MLB logo on back in reference to the pinstripes.

5)placement of numbers on back in reference to the pinstripes.

6)if any patches or memorial black armbands, then reference to pinstripes and/or edge of sleeves.

I make a complete checklist of all points of reference, even down to each bend of logos and numbers, each buttonhole, etc. Then start searching for photos. Getty Images is a great place to start because they will often have multiple photos from the same game. I copy each image from the specific jersey/player I'm researching and place in a folder. I'll then add any other photos I've located from other sources to the same folder.(making sure to rename each image with ref#'s, where it came from, & dates images were taken) Then I will go through the images checking each photo against my reference-point checklist. If the image has even one point of reference that doesn't match the jersey, I place it in a "no match" folder within my image folder. If I find images with some matching areas and some inconclusive areas, I place it in a "possible match" folder. If I find an image that matches all of the reference points, then I save it in a "match" folder and then target any other images from that same game.

If you are lucky enough to find multiple photos from the same game, repeat the process for each photo until you can be assured that each photo matches every reference point showing on each image.

Very time consuming and more often than not, very frustrating due to lack of photos or photo quality, however, I've photo matched my Giambi home pinstripe from 2006 that way.

Hope this helps.

[kingjammy24]

jason,

whether you're talking about photomatching a boardmark or pinstripes, you're ultimately referring to odds; specifically, the odds that a particular mark is truly unique. ie: what are the odds that a particular mark has not been duplicated?

when you see a boardmark, you feel it's a photomatch because the odds of another jersey with that exact boardmark is very slim. note though, that although the odds are slim, they're not zero. it is, afterall, technically possible for a hockey player to wear a jersey, get a boardmark, put on another jersey and get an identical boardmark in that identical location. technically possible but very slim. so slim that it qualifies as a photomatch. for this reason, i don't believe it's inherantly possible for any photomatch
to ever be a 100% guarantee. logically, there has to be room for the technical possibility that a particular mark was duplicated. that said, most collectors are fine with 99% odds.

"my guess is that logo's and patches may be sewn on slightly differently on each jersey so the pinstripes may appear in different locations from jersey to jersey. Is this correct?"

correct.

"If the above is correct then my next question would be....is this really a reliable way to photo match? Lets say a player wears 10 jerseys, all the same size and specifications, is it safe to say that 2 (maybe even more) of those jerseys may have pinstripes in identical positions using logo's and patches as references?"

you've missed a few things. it's not simply the logo and a patch. you've also got to match up the shoulder seams, side seams (if possible) and most importantly, the back nob and numbers. would i accept a photomatch that simply matched up the pinstripes with the front team logo? no. as you implied, it is safe to say that several jerseys use the pinstripes as guides for the logos and patches. the shoulder seams are better because i don't believe they serve as guides during the construction of the jersey. if you're cutting swaths of pinstriped fabric to construct jerseys, then the shoulder seam/pinstripe matchups are going to be more random and thus, more unique.

personally, when it comes to pinstripes i really like to see a match on the nob and rear numbers. with an average sized name and 2 digits, you're looking at possibly 20 different "intersections" between the pinstripes and various parts of the nob and numbers. in the context of photomatching, i feel that 15+ intersections are reliable. with each intersection you reduce the odds of a duplicate jersey having the same qualities.

as i mentioned, even with hockey jerseys, it's still about the odds. the odds of 2 jerseys having 4 identical boardmarks is pretty slim. however, don't you feel that the odds of 2 pinstriped jerseys having 20+ identical intersections is similarly slim?

rudy.

[Jason]

as i mentioned, even with hockey jerseys, it's still about the odds. the odds of 2 jerseys having 4 identical boardmarks is pretty slim. however, don't you feel that the odds of 2 pinstriped jerseys having 20+ identical intersections is similarly slim?

rudy.

Rudy,

You made good points to help me understand better, thank you.

But in response to your last statement... I would take a hockey jersey with 4 board marks over a pinstriped jersey with 20+ identical intersections any day! Especially since a pinstripe does not show game use..it just shows a pattern just like you would buy in a store.

The board marks are unique wear to the jersey made during a game. Intersecting pinstripes are part of a manufacturer pattern. Yes 20+ identical intersections would be pretty convincing assuming you can find other pictures of that same player wearing a pinstriped jersey without those 20+ matches..proving that your jersey matches another specific photo you have.

I have a high resolution getty image of a jersey I own and you can zoom in to several team repairs and match it stitch for stitch to my jersey, as well as other fabric pulls, stick/puck marks and loose threads. You could have a high resolution getty image of a bright white, clean pinstriped yankee jersey with 20+ pinstripe intersections and it still would not be as convincing to me.

[Jason]

also forgot to mention...

I see a lot of people on these forums showing "photo matches" by circling 2-5 pinstripes on the photo. i understand they are excited that they think they found a match, but the term photo match to me means a 100% match between a photo and the jersey in hand.

I can't help but laugh sometimes seeing people "matching" some many questionable items especially from a low resolution photo. Just because there is a blurry marking on a low res photo doesn't mean that makes it a photo match to a marking on an item.