Dr. Baker’s Vintage Golf Bag (Karsten IIIA)

The story, the specs, and the data.
Eighty-seven percent of a fitted bag at a fifth of the price.

This is my left-handed bag. The five clubs I bought for it cost $656.12, and everything else in it came from a $100 Woot set my father gifted me and a few years of thrift stores. Those five: a Cleveland Launcher XL 2 Draw driver, a Ping G410 3-hybrid, a Cleveland ZipCore XL 48° wedge, a Cleveland CBX ZipCore sand wedge, and a PXG Blackjack putter cut to fit a 6'5" wingspan on a 5'11" frame. The irons are Ping Karsten IIIAs from around 1980, a factory-matched left-handed set I found thrifting in Kentucky. Every purchase was bought used, checked against the actual spec (loft, flex, handedness), and negotiated. I never set foot in a fitting studio. I did it with a tape measure, a golf app, a Garmin R10 launch monitor, and a lot of eBay listings.

“Vintage where the design stopped changing. Modern where it didn’t. A shelf in the garage for everything else.”

This page is also a recipe. If you leave with one thing, leave with this:

“Anyone can build most of a modern bag out of the past. Find where golf stopped improving, buy vintage there for thrift-store money, spend real money only where the technology moved, and let the data grade your work.”

Watch first, read second

The idea

Golf equipment did not improve at the same rate everywhere. Drivers, sand wedges, and putters got dramatically better. Irons mostly did not. Karsten Solheim solved them in the 1960s with perimeter weighting, and that is still the idea behind every game-improvement iron sold today. So I spent money in the three slots where the technology moved and kept the 1980 irons. Against a $3,000 bag that costs me about 3 to 6 yards per full swing club, and nothing from 85 yards in, where most of my strokes happen.

What I spent

The bag as it stands

Fourteen clubs. The lofts on the Karstens are not guesses. I tracked down a full Karsten I spec sheet and the published Eye 2 specs, the generations right before and right after my set, and they agree within about a degree. The yards are now measured. Each row shows my average carry and total on good strikes from eight Garmin R10 range sessions (September 22 to October 4, 2026); a dagger (†) marks a number measured on the course with 18Birdies.

A note on the pitching wedge. Its loft worked out to essentially the same as the 9-iron’s, so a modern 48° took the slot and the Pinemeadow PW went to the shelf, along with the Karsten 4-iron, which had the same loft as the hybrid. Retired for good: a Karsten II 1-iron (16°, basically my 3-wood, a generation older than my set, and left-handed, which may make it the rarest thing on the shelf), the Karsten 3-iron (21°, one degree off my 5-wood), the Eye 2 and IIIA drivers, a Kroydon 9-iron from the hickory days, a Super Concorde 5, and a Lynx Predator 4-wood. Each one got replaced by a club in the bag that does the same job and launches.

What the IIIA actually is

Ping never wrote any of this down. Here is what I have pieced together from the classic-club forums, the used market, and the clubs themselves.

The dating. The Karsten I came out in the late 1960s, the Karsten II followed in the early 1970s, the Karsten III was on pro shop racks by 1976, and the Eye 2 arrived in 1982. My IIIAs sit in the gap, most likely 1978 to 1981. The garage shelf accidentally holds the timeline: a Karsten II 1-iron, the IIIA set, and an Eye 2 driver, three generations of Phoenix in one row.

The A. The A means left-handed, and the forums supplied the reason. An owner on GolfWRX who plays a IIIA set from sand wedge to 1-iron wrote that he has only ever seen them in left hand, and asked whether Ping made the original EYE in left hand at all, or whether the IIIA was what they sold left-handers instead. The used market answers him: right-handed Karsten IIIs are everywhere, right-handed IIIAs essentially do not exist. The III was the right-handed club; the IIIA was the same design built for the other side. For a few years, if you were a lefty who wanted Pings, this was the set.

The family. A right-handed Karsten III 1-iron listed online reads Toe-Heel Balance, 17-4 steel, red dot, the same stamps and steel on my heads with the words in the other order. The III sets ran through a sand wedge and so did the IIIA line, which means a left-handed IIIA sand wedge exists somewhere. It is on the thrift list.

The grooves. People who know Ping history ask about the Eye 2 square-groove lawsuit, the one that let John Daly play 1980s Eye 2 wedges on tour in 2010 under a grandfather clause. My irons are older than the argument. Square grooves arrived with the Eye 2 in 1984; the IIIA has ordinary 1970s V-grooves, never in dispute, never needing a loophole. The outlaws’ law-abiding older siblings. The Kentucky chapter is over, but the thrifting is not; I live in Los Angeles now and check the shelves everywhere I travel. If a thrift store anywhere ever hands me an Eye 2 wedge, it gets a plaque.

What they are worth. Karsten III singles trade for $25 to $45; a full left-handed set listed in the UK for about $190. The market treats them as good old clubs, not collectibles. Collectibles sit in cases. These play.

How I sourced the lofts

There is no factory spec sheet. An owner who called Ping with his serial number got a production year and nothing else. So I went the long way around. A full Karsten I sheet (1i 16, 2i 18, 3i 21, 4i 24, 5i 28, 6i 32, 7i 36, 8i 40, 9i 45, PW 50) and the published Eye 2 specs from 1982 (the sheet shows the long irons: 1i 16, 2i 18.5, 3i 21.5, 4i 25, with lies and lengths) bracket my set from either side, and where they overlap they barely differ. The IIIA came out between them, so 24 / 28 / 32 / 36 / 40 / 44 / 50 is my working ladder until a loft machine says otherwise. It also settled one thing for good: vintage lofts run 6 or 7 degrees weaker than a modern iron with the same number. Your 29° 7-iron is my 5-iron. Compare by loft.

The woods got their numbers from a forum owner who put a full Karsten wood set on a loft machine: 3-wood 15, 4-wood 19, 5-wood 22. So my 5-wood is about 22°, and my Pinemeadow 3-wood at 15° carries the exact loft a Karsten 3-wood did sixty years ago. Three-wood lofts never inflated; iron lofts inflated 5 to 7 degrees. The inflation was marketing, not engineering. His measured irons ran up to two degrees off the printed sheet, the 7-iron most. Old lofts wander, mine included; the machine gets the final word.

Table 1. The spec sheet nobody printed

Since nobody printed one, I built it. Lofts are the sourced ladder above. Lie and length are extrapolated from the Eye 2 sheet’s long irons (its 4-iron: 60° lie, 38.25"), which is how Ping built irons in that era, and the black dot on mine means standard. The modern column is a current Ping G440 set at Ping’s published standard specs, same number against same number. Shaft models are read from the surviving labels (weights per the vintage Ping shaft references, with the ZZ Lite at 115 grams confirmed in the compiled Ping shaft chart).

The loft gap has a shape: four degrees at the 4-iron, seven at the 7-iron, a degree and a half at the PW, zero at the wedges. Manufacturers strengthened the scoring irons hardest, because a longer 7-iron sells clubs, and could not touch the wedges without breaking the short game. The club my friends most want to compare is the one where the stamp lies the most.

Modern irons also run about a half to a full degree more upright and about half an inch longer, club for club. A 6'5" wingspan on a 5'11" frame wants flatter and shorter, and the Karstens are both. My fitted friends paid to have theirs bent. Mine came that way.

The shafts I read off the clubs. Labels survive on nearly every vintage Ping I own, and almost all say the same thing: Ping ZZ Lite, the company’s own 115-gram steel, on the woods, both shelf drivers, and most of the gamer irons. So the joke is confirmed rather than estimated: Ping’s 115-gram shafts were the lightweight option of 1980, up against everyone else’s 125-gram Dynamic Golds, and my set wears the light ones. Two clubs broke the pattern, and both are irons: the shelf 3-iron reads True Temper Dynamic S, and the gamer 5-iron’s faded label appears to read True Temper too. Two irons reshafted is the signature of a single repair visit somewhere in fifty years. The biggest single gap is still the 5-wood: 115 grams of steel against 65 of modern graphite, 50 grams heavier, and it went 174 off a tee anyway. The one shaft modern by weight is the driver’s Mitsubishi Tensei AV Blue 55, the same 55-gram class the big brands ship today.

Table 2. Same number, different truth

The tee-box comparison, same-number against same-number, with the loft gap that explains most of the difference and what each modern pick costs over what I paid. The picks are real models that come left-handed. The dagger on row 1: in MyGolfSpy’s 2024 test the G430 SFT and the XL 2 Draw carried the same distance to within a tenth of a yard; what the Ping buys is a tighter pattern, not yards.

Modern same-number irons fly 8 to 14 yards farther than mine and carry 4 to 7 degrees more loft to do it. At two yards a degree, the stamp accounts for nearly all of the gap; the engineering is worth 3 to 6 yards. Rows 1, 13, and 14 are the slots I paid for, each at or under the used market price. And a note on the new column: Ping cut G440 metalwood prices by about a quarter in August 2026, five months ahead of the next release cycle. The clubs the fitted crowd bought at full price are already marked down, which is the depreciation treadmill this bag was built to skip.

What the range and the course have said so far

Before the launch monitor. The CBX gave me three repeatable numbers almost by accident: 90 on a regular full swing, low 70s on a three-quarter, about 60 on a half. Early on my wedges stacked inside ten yards at normal tempo, and I suspected I swing the short wedge harder than the long one. Then the course numbers seemed to open the stack up: 18Birdies True Distance, built from ten-plus tracked shots per club, put the PW at 109, the gap wedge at 98, and the CBX at 90. They looked like proper gaps. The launch monitor would show that they were partly a mix of full and partial swings. Three clubs have each thrown one shot about 20 percent past the rest (149 with the 7-iron, 127 with the 9-iron, 116 with the gap wedge). I assumed I was tagging the wrong club. The swing analysis below says otherwise: I have two swings, and the long shots are the good one. One more check came from the backup set. The Pinemeadow 6/7 combo, about 30°, averaged 132 on normal swings (127 to 137), dead on my Karsten 5-iron where the ladder said it would land, and threw 152, 154, and 158 on three of eight, the same twenty percent jump. A club from a different maker, never used to build the ladder, confirmed both findings at once. And the driver joined the pattern in my latest nine: three drives at 190-197, two at 141-143, and one at 239, a new personal best. Same two swings, longest club in the bag. That nine, at Penmar in Los Angeles, also produced back-to-back pars, the second one saved by a chip-in from 39 yards with the CBX, which is not a data point, but I am counting it anyway. The wedge was bought for exactly that shot. And the 5-wood went 174 off a tee, forty yards past the 3-wood’s average. One swing is not a verdict, but it is a start.

Then the launch monitor arrived. Eight Garmin R10 sessions, about 640 swings, turned the targets in the bag table into measurements, and four things stood out.

The driver audition is over. Same session, alternating blocks, Chomp choked down against the Pinemeadow at full length: Chomp was about 4 mph faster off the face, carried about 4 yards farther, and curved about half as much. On the course it averaged 201 yards over 24 drives, against about 172 for the Pinemeadow before I changed my grip and 187 after. Fairways were a tie with my old setup at 32 percent, so what the Cleveland bought is distance and a straighter flight. It stays, cut to the length I choke it to.

My long irons bunch. On good strikes the 4-iron through 7-iron carry 123, 120, 116, and 113, three or four yards apart, where seven to nine is normal at my speed. The cause is not the irons. A 4-iron should launch lower and faster than a 6-iron; mine launches just as high (20.7° against 20.0°) at the same ball speed. I add loft at impact by bottoming out behind the ball, so my 4-iron plays like a 6-iron. That is why the 4-iron stays on the shelf and comes out only at the range, as a gauge for the fix.

My wedges overlap. The 9-iron, the Karsten W, and the 48 all carry 93 to 100 yards. The gaps the course logs showed came from mixing full and partial swings. Full swing, there is no room for another wedge between the 56 (82) and the 48 (93); the gaps get filled with half and three-quarter swings, so that is what I practice next.

The longer the club, the worse the strike and the farther left the miss. Wedges and irons are struck well about half the time, the hybrids about one time in four. The average miss runs 1 to 6 yards left with the short irons, 32 with the 3-hybrid, and over 40 with the drivers. Most of my iron misses (78 of 124) are fat, thin, or topped, the kind no iron, vintage or modern, forgives.

What the money actually buys

The honest part. Modern irons forgive a mishit better than mine do, and they launch higher, so they hold firm greens where mine land and run. I play for that by landing approaches five to ten yards short and letting them release. I am conceding the 13 percent I do not have so the 87 percent I got for a fifth of the price reads as fact.

What the swing is doing

I ran two swings through the 18Birdies analyzer, face on. One passed 4 of 11 checkpoints. The other passed 9. Same setup, same clubs, same golfer. Here is where they differ.

The bad swing is a chain: the lead arm bends at the top, the hips slide instead of turn, and the tempo goes with them. A bent lead arm shortens the swing and has to recover on the way down, which is the slightly fat, slightly thin strike that costs 15 to 20 percent of the distance. Hip sway moves the bottom of the swing and brings the club over the top with the face open, which is the slice. That chain explains the two sets of numbers in my logs. The jump is the same size on four different clubs now, roughly a club and a half, and on the combo it showed up three swings in eight. The useful part is that everything that differs already passes in the good swing. I am not learning a new move. I am learning to make the one I have more often. No club in anyone’s bag fixes that, including mine.

The update, two film sessions later. At the range I found the bad chain’s trigger, and it was a setup position, not a motion: my wrists were not close enough to my front leg at address. Moving them there presets forward shaft lean, and in one session the lead arm, hip sway, hip rotation, and tempo checkpoints all flipped to pass. Compression came with them, and the over-the-top move is absent from every swing filmed since. The analyzer’s down-the-line view then found what remains: spine angle and butt position at impact. Slow-motion overlay film, with the app drawing my address tush line and a box around my head, showed the fault is narrower than it first looked. Posture holds on the line through the top, the transition, and the delivery; in the final tenth of the downswing the hips surge off the line and the head exits its box, a late vertical jump at the ball, hunting speed. The head movement flagged in both original swings turned out to be this same fault measured at the top of the body, not a separate problem. So the whole remaining project is the last quarter-second: wall and basket drills where the rear stays back as the ball leaves, low wedges finishing waist-high, and a cue that fits on a scorecard: post, do not jump. The belt buckle turns. It does not lift.

The radar’s version. Front-on video synced to the R10 put a number on it. On my good swings my hands are over my lead thigh at impact and the shaft leans toward the target. On a topped 9-iron the club had already reached the ground behind the ball, hands back by my trail thigh, after a rushed transition (tempo 5.2 against 3.5 to 3.9 on the good ones). The R10 had been saying the same thing all along: a positive attack angle on every iron and wedge, meaning the bottom of the swing sits behind the ball. The fix I am working on is a shift, then a turn: weight onto my lead side early in the downswing, then rotate through. In the first session using it on every iron, the 9-iron’s attack angle dropped from about +5 to +2 and the block was the straightest I have hit. It is the same quarter-second as the cue above, seen from the other end: post, shift, and turn through with the hands ahead of the ball.

Where the bag actually ranks

By equipment I sit at solid intermediate: the scoring end is advanced, the long end is two boxed-set clubs and a 46-year-old Ping doing carefully assigned jobs, and the irons forgive like a beginner set and, until the strike improves, bunch at the long end. By management I am ahead of most advanced players, because the typical single-digit golfer knows his 7-iron number and guesses the rest, and I have a spec sheet. And by results I am a 22 with a 9-of-11 swing I make some of the time. The distance between my equipment tier and my scoring tier is the widest gap on this page, and no purchase closes it. The swing changes first, then the handicap, and the bag will still be fine when they do.

The replacement queue

Since people ask what I would upgrade, here is the whole bag ranked, first to last, with the trigger that would fire each move. The ordering logic: measured underperformance beats theoretical gaps, scoring-zone dollars beat long-game dollars, and clubs doing their jobs go to the back regardless of age.

Two things about the shape of this list. Rank 1 already fired, for $89.82, and the radar put rank 2 on hold, so the near-term upgrade path costs less than one fitted iron: cheap fixes where problems are measured, patience everywhere else. And notice what fires the triggers: a range audition, a handicap milestone, wear. Not one of them is a product launch. The queue runs on my data, and the market can run its treadmill without me.

The brand policy

When a trigger fires, the brand order is PXG, then Ping, then Cleveland, then whoever else fits. The order is part story and part arbitrage. PXG keeps the bag’s Marine thread intact, a Parsons iron set under a Parsons putter, and PXG has the steepest used-market depreciation in golf, which is a problem for people who buy new and a gift for people like me. Ping is the house the Karstens came from and has the deepest left-handed used market in the game, so the metalwood slots default there. Cleveland is for wedges, where it already lives in my bag and where nobody charges a lefty penalty. One rule sits above the list: fit beats brand, every time. Left-handed, right flex, right loft for the gap. If the PXG option means compromising any of those, the Ping option wins that day. The loyalty lives inside the method, not above it.

Where this started

For a few years I owned a house in Lexington, Kentucky, that backed up to the 18th hole of Andover Golf and Country Club, a private club that opened in 1990, a Clyde Johnston design playing just under 7,000 yards with a slope of 134. By the time I lived there it had closed. The fairway behind my house was a retired golf course, which is a strange and wonderful thing to have in a backyard. That is where the thrifting started. Between 2018 and 2023 I worked the thrift stores and estate sales around Lexington, and a town with that many closed clubs turns out to have a lot of good old golf equipment on shelves for a few dollars a club. The Karsten IIIAs came from one of those runs. So did the 5-wood, the Ping bag, the 1-iron, the Kroydon, and most of the museum shelf. A left-handed set of matched 1980 Pings is not something you go looking for. It is something you find when you are looking at everything.

And one of those finds closed a circle I did not notice until recently: the Ping cart bag still has an Andover member’s tag clipped to it, green horse logo, Todds Road address, est. 1989. When the club closed, its members’ gear scattered into the Lexington secondhand market, and one member’s bag found its way to the person living on the 18th green. I did not just thrift clubs while living on a dead course. I thrifted a piece of the course back.

The Pinemeadow set came earlier, a $100 boxed set off Woot in 2012 and a gift from my father, which is what I learned the game on. The five clubs I bought for this project are the only clubs I have ever paid real money for. Everything else in the bag was a deal, a discovery, or a gift from Dad.

These days the bag works out of Los Angeles. I play at Chester Washington, the first integrated golf course in Los Angeles County and the home course of Dr. Charlie Sifford, and at Maggie Hathaway, the par-3 named for the woman whose 1955 fight integrated it. A bag built on golf history gets to play on some.

The Gainesville chapter rides along too. My PhD and MS are from the University of Florida, and the long clubs travel in Gators colors: a “1” on the driver, which makes it Chomp, after the Gator Chomp, the “3” cover on the 3-wood, the “X” on the 4/5 hybrid, a Gators towel on the bag, and the Karsten 5-wood, the oldest and least replaceable head I own, under Albert, the Daphne’s alligator, the thickest padding in the inventory on the one club that has no replacement. And the delegation now has its own transport: a Ping Hoofer Vantage stand bag in Gators colors, the carry bag for the short-course days at Maggie Hathaway and Penmar, where seven or eight clubs walk and the vintage Ping bag stays home. Kentucky built this bag. Florida made it a doctor. Los Angeles is where it plays.

The skull on the putter

There is one purchase in this bag that was not made on specs. The PXG Blackjack debuted in October 2020 as the first release in PXG’s Battle Ready collection, a fully milled, tungsten-framed mallet that retailed at $525 built to spec. Mine wears the heel-shafted hosel, the toe-hang option of the four PXG offered, with a 370-gram standard head. That is the right hosel for an arcing stroke, which is the stroke I have. But the spec that sold me is on the sole. The putter carries what PXG calls the Darkness Insignia, a skull with the number 26. It honors Bob Parsons, the founder and CEO of PXG, and his service as a rifleman (0311) with the 26th Marine Regiment in Vietnam. I knew that before I bid, and it mattered.

My father, James Earl Baker, served twenty-one years in the United States Marine Corps and retired as a Gunnery Sergeant. I was born at Camp Pendleton, which is where the Marine Corps Test Unit that invented force reconnaissance (FORECON) doctrine was folded into the 1st Amphibious Reconnaissance Battalion in July 1957 and re-designated, the next day, as 1st Force Reconnaissance Company. My father taught me about Force Recon the way some fathers teach baseball. In his telling, they took the missions the Navy SEALs did not. The record is at least this generous: Recon Marines were developing clandestine insertion methods before the SEALs or the Green Berets existed. I carry that into work. The research lab I direct at USC is the Network Reconnaissance Lab, NetRecon, named for FORECON. We move critical data through networks that are intermittent and hostile to vulnerable people who need it. Reconnaissance, in the sense he meant it.

Checking the history for this page turned up a coincidence I did not go looking for. In 1966, First Battalion, 26th Marines was activated at Camp Pendleton, and a Force Recon platoon was assigned to deploy with it to Vietnam. The regiment on the sole of my putter was stood up at the base where I was born and went to war alongside the unit my lab is named after. I bought the putter because it fit my wingspan and my stroke, and it does. But I bid on it because of the skull. Every putt I make is on a club that honors a Marine, held by the son of one. And the first clubs I ever owned, the Woot set I learned the game on, were his gift too. The bag starts and ends with him. The putter is named Gunny. The 5-wood, the oldest survivor in the bag, is named Albert, after the Daphne’s alligator it wears and the University of Florida, where my PhD and the “Dr.” in this page’s title come from. The driver is named Chomp, after the Gator Chomp, for the Gators “1” cover that guards it. The 3-hybrid is named Vegeta, after the Super Saiyan Blue headcover from Dragon Ball Super; I am a lifelong Dragon Ball fan, and there is no better patron for a club bought to fix a gap than the prince who trains obsessively to close them. The names have moved more than the clubs. The first driver I bought, a G410, came with a plush parrot cover and was christened Barry, then spent a month held hostage in a FedEx facility over a label mismatch and never arrived. The refund came through, a G425 SFT took its slot and showed up right-handed, and the dealer swapped it for a Cleveland still in its factory wrap. Vegeta’s cover was too small for the new head, and so was Albert’s, so the driver went to a Gators “1” and became Chomp, Albert went home to the 5-wood, and Vegeta found the 3-hybrid. The parrot never made it out of Bloomington. Somewhere in a shipping warehouse there is a driver with a name and nowhere to go. This bag does not forget its casualties.

Why my bag tells the better story

1. Money. Your fitting bought about four yards a club at something like $100 to $200 a yard. My bag would sell today for about what I have in it. One of us made an investment. The other made a donation to a marketing department.

2. Scoring. From 85 yards in, our bags are the same technology: a current sand wedge and a premium mallet. A 22 handicap leaks eight to twelve strokes inside 100 yards. That is where the first money I spent on this project went. Your fitted long irons will look wonderful on the two par-5s where they matter.

3. Fitting. I got fit too. Putter length to a 6'5" wingspan, driver flex to a swing speed my radar measured at about 81 mph, every carry measured on a launch monitor, and every job assigned from a logged baseline. Then I measured the swing itself and found the real problem, and it was not equipment. The analyzer says I already own a 9 of 11 swing. I just do not make it every time. Nothing on your side of the table fixes that either. Fitting is measuring. It does not require a fitting studio and a latte.

Across three driver hunts I vetted more than twenty listings; the rejects included stiff, extra stiff, senior, cut-down, damaged, and secretly right-handed, and one of them made it all the way to my door before the hand showed. The one correct club was worth the month.

4. The story. A factory-matched Karsten IIIA set from around 1980, the model Ping built for left-handers when the flagship was not offered in left hand, rare enough that even Ping has no spec sheet for it, found while living on the 18th hole of a closed Kentucky course, under a Cleveland driver forty-odd years younger, beside a skull-stamped putter named Gunny after the Gunnery Sergeant who raised me, a driver named Chomp after the Gator Chomp, a 1980 5-wood named Albert after a Gainesville doctorate, and a 3-hybrid named Vegeta after a Super Saiyan, all succeeding a driver named Barry that FedEx never released. Your bag matches. Mine has a plot, and a hostage subplot. Nobody at the 19th hole asks to hear the story of an invoice.

“I own about 87 percent of your bag’s performance for about 20 percent of its price, all of its short game, and the only origin story worth telling. See you on the tee. I will be the one asking what you really hit into that last green.”

Karsten lofts are sourced from the Karsten I spec sheet (see also this breakdown and this measured set) and the Eye 2 specs, wood lofts from a machine-measured Karsten wood set, and all are still pending a loft machine of my own. Measured yards are from a Garmin R10 on the range and 18Birdies on the course. Prices are from live listings and completed sales.