Writing
Notes on measuring golf courses
Longer pieces are published under my own name rather than here. They are free, there is no schedule, and a new one appears when there is something worth saying. Newest first.
October 2026 · for boards and green committees
The Second Golden Age Has No Specification
American golf is a century past its first building boom and deep into a second one, putting those courses back at $10-20 million a project. The construction specification is rigorous: rootzone depth, laboratory-tested sand, percolation rates, all of it verified. What no document in the file states is what the finished golf course will do: how hard, for whom, by how much. Oakland Hills, Medinah No. 3 and the Old Course, and what a performance specification would have said before the first bunker was dug.
Read the essayA member facing an $11 million dues increase had no way to know whether his golf course would come back easier, harder, or the same.
Golf architecture has the materials specification, in impressive detail. For the other half of the pair it has adjectives.
September 2026 · for superintendents
Measure the golf, not just the grass
A superintendent runs one rigorous, instrumented system for turf health, and one for playability that runs on tradition. Green speed, rough height, aerification timing and firmness all change how the course plays, and almost none of it is decided from measurement of the golf. What the scoring record can price, what it cannot, and the 4 things worth logging from Monday.
Read the essayAsk what a Stimpmeter reading of 11 costs the membership, in strokes, compared with 10, and there is no answer.
Not because the question is stupid. Because nobody has ever been able to produce the number.
August 2026
Golf Measures Everything Except the Golf Course
Golf courses change gradually enough that nobody notices, and the system that describes them is a prediction made on foot once a decade rather than a reading of the scores that followed. Horse racing, baseball, cricket, tennis and track and field all measure the surface their sport is played on. This is an argument that golf should too, and what it costs clubs that it doesn’t.
Read the essaySomewhere in your clubhouse there is a board with a course record on it. A name, a number, a year.
Nobody knows what the wind, greens and rough were doing that day.
June 2026
Which U.S. Open venues are actually the hardest?
Winning score tells you about one player's week, not the golf course. Adjusting for era and field strength across 5 independent datasets produces a four-tier ranking that is tested rather than asserted. Merion tops the modern era at under 7,000 yards. Oakmont played 5.47 strokes harder than its co-host on the same day without championship setup — and when the Amateur reached Merion in 2026, it played only 0.81 strokes harder than its own co-host, which is the difference between a hard course and a hard setup.
Read the essayLength is not difficulty. Neither, on its own, is architecture.
Hugh Wilson gave Merion greens small enough to humble anyone without a yard of help. The 2026 Amateur finally measured how much of its championship severity is Wilson, and how much is the USGA.
Papers
The essays above make the argument. These are the evidence behind it: 4 papers posted open access, each with its own DOI, with full method, data and code deposited openly so that any number in them can be checked or contradicted.
The longest course played the easiest
Across seventeen U.S. Opens, course length accounts for about nine per cent of the variation in scoring, and the relationship runs slightly the wrong way. Erin Hills, at 7,741 yards the longest venue of the modern era, produced the easiest championship; Merion, 745 yards shorter, produced the hardest.
Cite this paper
BibTeX@misc{mohiuddin2026usopen,
author = {Mohiuddin, Junaid},
title = {Measuring Championship Venue Difficulty in US Open Golf},
year = {2026},
publisher = {SportRxiv},
doi = {10.51224/SportRxiv.947},
url = {https://doi.org/10.51224/SportRxiv.947}
}
APAMohiuddin, J. (2026, July 6). Measuring championship venue difficulty in US Open golf [Preprint]. SportRxiv. https://doi.org/10.51224/SportRxiv.947Chicago
Mohiuddin, Junaid. “Measuring Championship Venue Difficulty in US Open Golf.” Preprint, SportRxiv, July 6, 2026. https://doi.org/10.51224/SportRxiv.947.
Two strokes apart, in the same week, every year
The Alfred Dunhill Links rotates a single field across three courses within one championship, which holds the field constant by design. Carnoustie played between 1.37 and 2.44 strokes harder than the Old Course in each of five consecutive years, averaging 1.85.
Cite this paper
BibTeX@misc{mohiuddin2026open,
author = {Mohiuddin, Junaid},
title = {Measuring Venue Difficulty in the Open Championship},
year = {2026},
publisher = {SportRxiv},
doi = {10.51224/SportRxiv.1007},
url = {https://doi.org/10.51224/SportRxiv.1007}
}
APAMohiuddin, J. (2026, September 10). Measuring venue difficulty in the Open Championship [Preprint]. SportRxiv. https://doi.org/10.51224/SportRxiv.1007Chicago
Mohiuddin, Junaid. “Measuring Venue Difficulty in the Open Championship.” Preprint, SportRxiv, September 10, 2026. https://doi.org/10.51224/SportRxiv.1007.
The most controlled championship is the least repeatable
Measured as pooled within-venue variance in field scoring average — how closely a venue reproduces its own result on a return visit — the PGA Championship sits at 0.160 against the U.S. Open’s 1.360. The championship with the most deliberate setup intervention is the least reproducible of the three, by a factor of eight.
Cite this paper
BibTeX@misc{mohiuddin2026pga,
author = {Mohiuddin, Junaid},
title = {Convergence and Reproducibility in Championship Venue Difficulty},
year = {2026},
publisher = {SportRxiv},
doi = {10.51224/SportRxiv.1064},
url = {https://doi.org/10.51224/SportRxiv.1064}
}
APAMohiuddin, J. (2026, September 11). Convergence and reproducibility in championship venue difficulty: Evidence from the PGA Championship, 2010-2026 [Preprint]. SportRxiv. https://doi.org/10.51224/SportRxiv.1064Chicago
Mohiuddin, Junaid. “Convergence and Reproducibility in Championship Venue Difficulty.” Preprint, SportRxiv, September 11, 2026. https://doi.org/10.51224/SportRxiv.1064.
Stated before the fact. The paper’s decomposition yields a forecast for the 2027 U.S. Open at Pebble Beach:+3.0 strokes over parfor the field scoring average, with an 80% predictive interval of+1.7 to +4.6.The mean of the 3 previous Opens there is +3.2. If June 2027 lands below +1.7, the USGA has changed how it sets up Pebble Beach; if it lands above +4.6, the shoreline weather will say why. A registered version with the weather term added will be posted in May 2027. The forecast was placed on record before Pebble Beach and Gil Hanse announced the alterations made ahead of the championship; the paper expects that work to add roughly +0.2 to +0.5 strokes, which is inside the interval. Scored leave-one-out against the three Opens already played, the method halves the error of a baseline-only forecast.
Cite this paper
BibTeX@misc{mohiuddin2026fixedvenues,
author = {Mohiuddin, Junaid},
title = {Fixed Venues: How Much a Championship Course Moves When Nothing Changes},
year = {2026},
note = {Preprint v2.1},
doi = {10.17605/OSF.IO/3HGJ2},
url = {https://doi.org/10.17605/OSF.IO/3HGJ2}
}
APAMohiuddin, J. (2026). Fixed venues: How much a championship course moves when nothing changes (v2.1) [Preprint]. FogBow LLC. https://doi.org/10.17605/OSF.IO/3HGJ2Chicago
Mohiuddin, Junaid. “Fixed Venues: How Much a Championship Course Moves When Nothing Changes.” Preprint v2.1, FogBow LLC, October 20, 2026. https://doi.org/10.17605/OSF.IO/3HGJ2.