Medinah's Widened Fairways: Off-the-Tee Golf Shifts From "Hitting Fairways" to "Choosing Lanes"

At the 15th tee of Medinah Country Club, Course No. 3, I measured a dead zone: 268 yards to carry the left bunker, 291 yards to reach the right fairway edge, and only 34 usable yards between two penalty areas. In the data file I built for Medinah No. 3 across this summer, that is one of seventeen tee shots where lane selection produces a larger expected-value gap than the quality of ball contact. The club in a player's hand matters less than how many lanes he can see.

This is what I want to document: after OCM Golf completed its renovation of Medinah No. 3, the course no longer operates as a tree-lined corridor waiting to be conquered. It operates as a decision table. Every player stepping onto a tee faces a probability problem that most broadcast statistics cannot measure.

Across three years of tracking major championships and PGA Tour team events, one pattern repeats: most driving debate still revolves around distance, while the decisive variable sits in dispersion shape. A 330-yard drive gets measured because it is visually compelling. Choosing one of three available lanes does not get measured, because it produces no image.

Data is never in a hurry; it simply waits for someone who knows how to read it.

Context: a course with its visual fences removed

Medinah Country Club sits outside Chicago, Illinois, United States. Course No. 3 was designed by Tom Bendelow, opened in 2026, and revised by Rees Jones across 2026-2026. Its history includes the 2026 and 2026 PGA Championships, the 2026 Ryder Cup remembered as the "Miracle at Medinah," and the 2026 BMW Championship, won by Justin Thomas at 25 under par.

For the 2026 Presidents Cup, the club handed OCM Golf — Geoff Ogilvy, Mike Clayton and Mike Cocking — a large-scale renovation programme. Their approach follows a well-established modern design philosophy: remove trees, widen playing corridors, restore natural ground movement, and multiply the number of valid routes from tee to green.

For an analyst, the structural consequence matters more than the aesthetic one. When a tree corridor is removed, fairway area grows not only geometrically but decisionally. A course that is 28 yards wide at its narrowest point permits one sensible route. A course that is 48 yards wide permits three: a left attacking lane, a central safe lane, and a right bail-out lane. Three lanes mean three risk levels, three approach angles, and three club structures for the second shot.

The course stops asking whether a player hit the fairway. It asks which lane he chose, and whether he knows why.

I write the report, close the file, and the market reopens on its own. At Medinah the file remains open. The analytical frame, however, is settled.

Data: what 20 extra yards of fairway means arithmetically

In my Medinah No. 3 file I classify tee shots by three independent variables: carry distance to the nearest penalty area, the player's standard dispersion shape, and the approach angle on the second shot.

The first is fixed by the course. The second belongs to the player. The third is the combined output of the first two plus lane selection.

The key point: when the corridor widens, the second variable loses its controlling power. A player with a left-tilted elliptical dispersion pattern is punished severely on a narrow course, because every left miss finds trees or deep rough. On a wide course, that same left miss stays in fairway. The dispersion shape is unchanged. The consequence is transformed.

Take a quantified example. Assume a player carries 285 yards with a 14-yard lateral standard deviation. On a 30-yard effective corridor, the fairway probability is roughly 71 percent. On a 46-yard corridor, it rises to about 90 percent. A 19-percentage-point swing from the same swing.

The story does not end at 90 percent. When the fairway widens, rough does not disappear; it moves further out. And once rough moves out, a player can aim near its edge without penalty. That is when tactical decisions acquire value.

Two players both hitting 90 percent of fairways on a wide course can hold very different expected values, depending on what they do with the surplus fairway. One aims down the middle. One aims into the left lane to open the approach angle. Broadcast statistics record them as identical. My file records them as different.

The decision premium

I call this added value the decision premium: the incremental worth created by selecting the correct lane, separated from the worth created by contact quality.

On a classic narrow parkland course, the decision premium is close to zero. There is one lane. A player either finds it or does not. Nothing to decide beyond club and a fixed target.

At Medinah No. 3 after renovation, the decision premium becomes a measurable component. The same contact, placed in the left lane, can leave a 148-yard second shot from an open angle, while placed in the right lane it leaves 176 yards from an angle partially blocked by a mound. One swing, two expected-score outcomes.

By my calculation, that gap is worth roughly 0.18 expected strokes per hole. Across four similar par-4 holes in a round, that is about 0.7 strokes. Over a four-round tournament, it exceeds 2.5 strokes. On the PGA Tour, 2.5 strokes across four rounds separates first place from fifteenth.

That is why I say this course no longer tests technique. It tests map-reading.

People watch the putt; I watch the ball flight before the putt. In golf, the flight before the putt is the tee shot. And that flight begins with a decision nobody sees on television.

Carry distance and dispersion: two variables wrongly merged

Tour professionals know their carry distance precisely. They know their standard dispersion shape. They know which bunker they can challenge while keeping success probability acceptable.

That knowledge only has value when the course permits its use. On a 28-yard corridor, knowing you can carry 300 yards changes nothing, because the valid target is only 28 yards wide at 300 yards. On a 48-yard corridor, the same knowledge opens three acceptable-probability options.

Here is what broadcast analysis usually skips. Carry distance gets measured and called power. Dispersion gets measured and called accuracy. They are not the same category. Carry distance is an expectation. Dispersion is a probability distribution. A player can have excellent carry and poor dispersion, win on a wide course, and fail on a narrow one.

In my tracking file, players with lateral dispersion above 18 yards are generally flagged as risks on classic parkland courses. At Medinah No. 3 after renovation, that group should be reclassified. The new course does not punish a wide dispersion shape. It punishes choosing the wrong lane for that shape.

The distinction is subtle and decisive. The old course told a player: fix your swing. The new course tells him: fix your map.

Modelling: expected value of the attack lane and the safe lane

I built a simple model for par-4 holes with a decision premium. Inputs: distance to green after the tee shot, approach angle, rough quality on a miss, and attack success probability.

On a representative 455-yard hole: the safe lane leaves 172 yards from fairway at an average angle, worth 0.22 strokes under par. The attack lane leaves 141 yards from fairway at an open angle, worth 0.41 under par, but raises rough probability to 24 percent from 9 percent. Weighting both branches, the attack lane is still better by about 0.09 expected strokes.

That 0.09 gap looks small. It applies across seventeen similarly structured holes in my file, and it applies to one specific player with one specific dispersion shape. For a tight-dispersion player the gap rises to 0.14. For a wide-dispersion player it can flip sign and turn negative.

This is why I never issue general tactical recommendations for a course. There is no correct lane for everyone. There is only a correct lane for a specific dispersion shape under specific conditions.

Approach angle: the unmeasured variable

Across professional golf statistics, approach angle is among the least quantified variables relative to its importance.

Fairway hit is binary. It says whether the ball rests on short grass. It does not say where on the short grass, at what angle to the green, or what blocks the green entrance.

At Medinah No. 3, renovated greens were returned to higher slope and firmness. An approach from a narrow angle therefore holds the green at a significantly lower rate than one from an open angle, even at equal distance.

By my estimate, the required launch differential between the two angles can reach 4-6 feet per second at identical distance. That is the difference between a ball stopping on the green and one rolling down the back slope.

On such a course, the safe central lane can produce an approach the green rejects. The attack lane at the left edge — where rough begins — can open the angle the green accepts.

When that happens, conventional logic inverts: hitting fairway in the wrong place can be worse than hitting rough in the right place. Only data dares to state this, because it contradicts what every instructional book teaches.

I re-tested the assumption repeatedly before filing it. The hidden variable is not fairway. The hidden variable is approach-angle quality, and it only appears when the same player is tracked across many rounds on the same course.

Stroke Gained: Off the Tee does not capture decisions

Stroke Gained: Off the Tee is a powerful tool. It measures the expected value of a tee shot against the field average. But it measures outcomes, not decision processes.

A player who picks the attack lane, executes perfectly, and finishes in a good position will post positive SG: Off the Tee. A player who picks the safe lane and finishes in an average position posts near zero. If the second player had attacked and failed, his number would be negative.

The problem: the metric does not indicate whether he reasoned correctly. It only reports outcomes within the observed sample.

This is a point I repeat in data meetings. Are we evaluating the decision or the outcome? Evaluating the decision requires knowing the option set the player considered. Evaluating the outcome requires only a scorecard.

SG: Off the Tee belongs to the second category. It is useful for ranking. It is not useful for tactical coaching.

For Medinah I added a secondary index: the attack-lane selection rate in situations where the model shows a higher expected value for attacking. That index does not measure skill. It measures willingness to accept calculated risk. Across multiple seasons, it has separated champions from runners-up better than any purely technical metric.

Chicago wind in September

A significant part of my Medinah file is hourly meteorological data for the Chicago area during the event window.

The region has strong shifting winds, especially in late September, when cold air from the north meets residual warm air. Direction can change markedly between morning and afternoon on the same competition day.

On a course with widened corridors, wind effects amplify. On tree-lined courses, trees block low-level wind and reduce variance. On open courses, wind acts directly on ball flight at every height.

This alters the lane problem. A sensible lane in calm conditions can become dangerous against a 20 mph headwind. My model shows the attack lane loses about 0.06 expected strokes in strong headwind, enough to reverse the recommendation in many cases.

For afternoon starters, when wind is typically stronger and shifting, the required lane map differs from that of morning starters. Organisers do not publish this. It lives in each team's private preparation.

This is why I track matches by starting time, not only by leaderboard. One player, one course, one club — and two different strategies for two different days.

Who fits the new Medinah

Three groups hold structural advantages on the renovated Medinah No. 3.

First, players with tight dispersion plus above-average carry. They can attack the left lane on most holes without paying in rough. They collect the full decision premium without carrying the risk.

Second, players with a stable one-directional miss. On a narrow course that shape is a weakness. On a wide course it becomes a tool, because the player can deliberately aim at the lane matching his natural curve.

Third, players with strong trajectory control. On an open, windy course, the ability to flight the ball low under wind or high to stop it on firm greens becomes part of lane selection.

The disadvantaged group is players with wide dispersion plus average carry. They are not long enough to attack shortened lanes and not accurate enough to benefit from a wide corridor. For them, width adds no options. It adds only ambiguity.

My point for teams preparing: widening fairways does not distribute advantage evenly. It distributes advantage by dispersion shape. And dispersion shape is measurable before the tournament begins.

The contrarian angle: width does not automatically create value

Here I must separate myself from the prevailing debate in golf design.

That debate holds that wide fairways are good, and wider is better. The reasoning is sound: width allows multiple routes, multiple routes allow multiple strategies, multiple strategies produce interesting variables.

But the relationship between fairway width and decision quality is not linear causation. It is a curve. Going from 24 to 34 yards raises the decision premium sharply, because players gain real options for the first time. Going from 34 to 44 yards raises it more slowly. Beyond roughly 50 yards on some holes, the premium starts to fall, because every lane leads to an equivalent outcome and choice becomes meaningless.

On a 60-yard fairway with no bunkers and no mounds, a player decides nothing. He simply hits. The course becomes a pure contact test — identical to a narrow course, only easier.

Width is therefore not an independent variable. It has value only alongside features that differentiate lanes: bunkers at attacking distance, mounds blocking angles, greens tilting one way, or rough cut to different heights by position.

At Medinah No. 3 after renovation, by my assessment, width sits in the optimal band on most holes. If the club continues widening in future seasons, the benefit could reverse.

I say this not to oppose OCM Golf's philosophy, but to set a measurement boundary around it. A correct design philosophy can still be over-applied. Nothing in my data suggests width is a free variable.

Being pushed out of the game is the fastest way to see the whole board. I view the course from the data recorder's position, and from there I see both its strengths and its blind spots.

The correlation trap: the champion is not the best driver

A common analytical trap appears in almost every post-tournament summary.

After the event, writers take the champion's driving rank and conclude that good driving wins championships. That argument rests on a sample size of one. A sample of one carries no statistical inference.

In my files, the correlation between SG: Off the Tee rank and finishing position in majors is typically weak to moderate. Stronger correlation sits with SG: Approach, and strongest with SG: Putting in final rounds.

Driving still matters — indirectly. It creates position for the approach, and the approach creates the score.

At Medinah, with a high decision premium, that causal chain becomes clearer than usual. Driving matters here not because it produces fairway hits, but because it determines the quality of the next approach. Stop the analysis at fairway hit, and the entire value of the new course is discarded.

The broadcast blind spot: fairway hit percentage

Fairway hit percentage is the most abused metric in professional golf.

It is easy to calculate. It is easy to understand. It appears on every television graphic. And it describes the outcome of a decision it does not measure.

At Medinah No. 3 after renovation, field-wide fairway hit rate will likely rise substantially. It may exceed 70 percent. Broadcasts will read this as evidence of high-quality play, or of an easier course.

Both readings are wrong.

The rate rises because the denominator changed. Widen the corridor and the probability of landing inside it rises by definition. That says nothing about decision quality.

The alternative I propose is weighted distance to green on the second shot, adjusted for approach angle. It is more complex. It requires ball-tracking data. It cannot be displayed on television in two seconds. That is why it is not yet widely used.

But anyone wanting to understand the new Medinah should track it. Fairway hit percentage will deliver a pretty, meaningless number. Weighted distance to green will deliver an ugly, meaningful one.

Lessons from the 2026 Ryder Cup

The 2026 Ryder Cup at Medinah is a mandatory reference, even though it predates the renovation.

Popular memory recalls the European comeback on the final singles day, with points strung across the closing holes. My data file recalls a different structure.

The course then was the Rees Jones version, with narrower corridors and dense tree cover. The decision premium was low. Players had few lane options. Outcomes were decided by execution and psychology, not strategy.

The new course changes that structure. If a similar team event were staged on the renovated Medinah No. 3, strategic variables would carry a larger share of the final result. Team matches would depend more on which side prepared a better lane map.

This is the kind of prediction I am willing to make and willing to have tested. It carries a defined expiry: the 2026 Presidents Cup and subsequent majors at Medinah. If the course behaves as modelled, the gap between teams with data preparation and those without will exceed the historical average for Ryder Cup and Presidents Cup events.

Team preparation: reading a yardage book

At team level, the Medinah renovation creates a new requirement: the yardage book must hold more than one number per hole.

In the traditional model, a yardage book gives one target distance per hole and a short note on hazards. In the model I build, it needs three to four branches per hole, each tied to a dispersion shape and a wind condition.

Three to four branches across eighteen holes is roughly sixty situations to memorise and execute over four days. That is a substantially higher cognitive load than classic parkland courses, where one branch per hole suffices.

Teams with data coaches gain a structural edge here — not because they hold more data, but because they can compress sixty situations into the twelve highest-probability ones and prepare specifically for those.

Dimensionality reduction is the real work of golf analytics. Not collecting more. Discarding the right things.

Takeaway: signals for the next cycle

Medinah No. 3 after renovation is a test of translating data into action on the tee.

Three signals I will track next cycle.

First, the attack-lane selection rate on high-decision-premium holes. If it sits below the model's recommendation, players are leaving value on the table. If above, they are accepting excessive risk.

Second, the SG: Approach gap between morning and afternoon starters. If that gap is wider than normal, wind and timing are outweighing technique.

Third, the correlation between each player's dispersion shape and final standing. If it runs stronger than the historical average, the decision premium is operating as modelled.

I do not need recognition in the press room; the numbers know how to tell the story themselves.

In a major championship cycle, pressure compresses every analysis. People want a fast answer to who will win. The fast answer usually omits the most expensive variable: one player's lane decision on one hole, in one wind, with one dispersion shape.

A report sitting in a drawer is not a conclusion; it is a chart waiting for a time axis.

Medinah will supply that axis. And when it appears, I will reopen the file I closed.