Five Bermuda Picks, Zero Prices: The Column That Fails the Ledger Test

I opened Brady Kannon's 2026 Bermuda Championship picks column on GOLF.com looking for a ledger. I found five names: Andrew Novak, Eric Cole, David Ford, Lee Hodges, Taylor Montgomery. I found course-fit reasoning, a list of correlated courses, a weather forecast. I found no prices. Five picks, one SG: Approach rank, zero SG: Putting figures, zero odds: those four numbers are the column's real identity.

In 2026, I carried a tablet for four rounds as a walking scorer at the Asian Tour's Bangladesh Open at Kurmitola. Nobody asked me to keep the file. I kept it anyway: more than 1,100 shot records. From that week, every golf piece I wrote began with at least three numbers: field size, purse, scoring average. A picks column can carry those three. Without a fourth number, the account stays incomplete: the price. Kannon's column has five names and a course-fit argument. The number that should sit beside every pick, the odds, is nowhere in the text.

A picks column without prices can prove narrative coherence. It cannot prove expected value. I do not trust a pick until it survives the ledger test. The first question is simple: what is the price, and what remains at that price over a long sample? The second question: is the metric justifying the pick actually measured, or merely named? Kannon's column breaks at the second question.

Port Royal Golf Course is a par 71, roughly 6,800 yards, Bermudagrass, low rough, large greens. The forecast calls for rain on Tuesday and Sunday, with wind at 15–30+ mph. The 2026 Butterfield Bermuda Championship is the fourth of eight FedExCup Fall events and the second of four consecutive non-US events. The field is not strong. It is a fall-series survival tier. Course fit matters here. Without a price, the fit argument remains incomplete. Kannon maps Port Royal to TPC San Antonio, TPC Summerlin, Pebble Beach, Waialae, Sea Island, PGA National and Colonial. Shared turf, wind exposure and rough height make the heuristic reasonable. There is no regression, no SG-by-course-type normalization, no field-strength adjustment. Correlation without a quantified fit model is a narrative, not a model.

The broadcast schedule is the quiet engine under every rights valuation; the quiet engine under a picks column is the price. Kannon says wind and rain make SG: Approach the first-order skill. The logic is clear: wind demands flight control, rain makes greens harder to hold. Across five picks, he supplies exactly one SG: Approach rank: Hodges, 24th. On the metric he places first, four of his five picks carry no rank at all.

Evidence density shifts from pick to pick, and it is measurable. Cole has three quantified metrics: third in 3-Putt Avoidance, 12th in proximity from 100–125 yards, second from 125–150. He also has eight correlated-course results: 14th and 26th at San Antonio, third and 15th at Sea Island, 14th and 15th at Pebble, fifth and 13th at Sony. No current-form figure is cited. Cole is the pick with the deepest historical correlation and a blind spot on present form.

Montgomery is the weakest case. No quantified metric. No recent-form result. The sentence in his favour, that he is one of the best putters in the world, is an evaluation, not a measurement. No SG: Putting figure appears anywhere. Cole and Montgomery are both putter-reputation picks. Reputation is not data. It is an old account that does not price itself into a new week.

Ford offers the clearest evidence. He is a rookie who turned pro in 2026 and finished No. 1 in PGA Tour University. He ranks No. 1 in this field for driving accuracy, No. 1 in 3-Putt Avoidance, third in Scrambling, and third at Utah three weeks ago. His amateur wins came at Sea Island and Colonial, with a Junior Players title at TPC Sawgrass. The professional sample is thin, but the metrics and recent form both exist.

Hodges fits Bermudagrass. The Alabama player has finished sixth and 10th at Waialae, sixth at the 2026 Valero, and fourth and fifth at Sea Island over the last two years. In this column his SG: Approach rank is 24th, driving accuracy 14th, proximity from 125–150 yards 15th. Recent form shows third at the John Deere and 13th at Utah. The evidence is balanced, but his top-line metric sits mid-field.

Novak has fewer numbers and a longer story. His only quantified metric is 14th in proximity from 125–150 yards. His correlated-course record is deep: 14th at San Antonio this season, third in 2026, ninth in 2026; top-15 at Sony and Pebble; seventh at Sea Island last fall; runner-up in Bermuda the prior year. He was seventh and 17th in his last two starts. History and form make him the most balanced pick in the set.

A methodological trap sits inside field-relative ranks. No. 1 in this field does not mean No. 1 on tour. In a weak fall field, No. 1 driving accuracy may be tour-average in a signature event. Kannon is honest when he writes in this field. The reader who treats the rank as an absolute number reads it wrong. Without a field-strength adjustment, rank comparisons remain incomplete.

The weather thesis and the putting thesis also pull against each other. Kannon forecasts wind and rain, then concludes the event may become a putting contest among players who solve the front end. Mechanically, wind suppresses greens in regulation and elevates scrambling. Rain softens greens, suppresses three-putts and reduces the discriminating power of 3-Putt Avoidance. Wind and rain pull scrambling and three-putt avoidance in opposite directions. The column does not resolve which regime dominates.

One part of Kannon's argument is strong. Large Bermudagrass greens mean the field will three-putt more than usual, which raises the value of 3-Putt Avoidance rank this week. The column's strongest technical argument is 3-Putt Avoidance, not SG: Approach, and that does not match its own stated priority.

The contrarian angle is this: the column's problem is not a wrong pick. It is a wrong product. Kannon is a SportsGrid host and CBS Sportsline analyst, with a commercial interest in betting-content engagement, and the column discloses the affiliation. Disclosure is good. It does not fill the price column. A free picks column sells narrative; prices reduce clicks, stories raise them. This column works well inside that model. It has names, course fit and weather. A pick can be valued only when its price is known. Without a price, the reader learns which story is coherent, not which pick holds value over time.

Time sensitivity is brutal. The column's value goes to zero at the first tee. The last-20-rounds window is itself stale, tied to each player's most recent starts and built against weak fall fields. Small sample, non-representative opposition, no major-championship conditions: put those three together and the rank stops being a permanent truth.

Bangladesh golf has lived with this price gap for longer. The BPGA circuit has one-week events and 51 silent weeks; there is no verified domestic live telecast, no odds, no price ledger. In that setting, No. 1 in this field means even less, because the field is small and the course pool is smaller. My week as a walking scorer at Kurmitola taught me to know the market before trusting the rank. The column's empty fairway is the price column: full of story, empty of number.

On Monday, an operator's checklist is short. Put a price beside every pick. Adjust ranks for field strength. Supply SG: Putting numbers. Run separate models for wind and rain. Publish the date and the data window. Set a minimum threshold for current form. Without those six, a picks column is a course-fit essay.

Bermuda week will end and the scoreboard will be wiped. The ledger remains, if someone keeps it. Five names and zero prices leave the reader with one question: if a pick cannot show its price, why should anyone trust its story?