Monthly releases
The index measures what one airline mile or hotel point buys. For each trip in a fixed basket, the cash price divided by the points price gives the value of a point. Each month I compare every trip with the same trip a month earlier, average those changes, and chain the result onto the month before, the way a consumer price index is built. In effect it’s an inflation measure for a currency that companies issue themselves.
This page is the index over time: the series, every monthly release, how much the latest month’s number depends on the method, which programs moved it, and how much of each month’s collection made it in. The rules and the arithmetic are in the research summary.
| Index level | 92.4August 2026 = 100 · three months collected |
|---|---|
| Change, September to October | −10.7%151 trips priced in both months |
| What is measured | Value of one pointCash price ÷ points price, in US cents |
| How it is averaged | Geometric mean, chainedSame trips, same booking distance, month to month |
| The basket | 88 trips × 2 booking distances27 programs; about 60 and 180 days ahead |
| Collected | Oct 16 – Oct 21, 2026By hand. 528 searches and 849 screenshots so far |
- Sep 4, 2026. IHG One Rewards → Air India Maharaja Club: partner added; collection rule 10 clarified.
- Sep 9, 2026. Hyatt World of Hyatt + Delta Skymiles: partner added.
- Sep 15, 2026. Wyndham Rewards: policy change takes effect.
- Oct 1, 2026. Chase UR → Hyatt World of Hyatt: transfer ratio change takes effect.
- Oct 15, 2026. Hyatt World of Hyatt: policy change.
- Dec 17, 2026. Korean Air SKYPASS: policy change takes effect.
- Dec 31, 2026. Hyatt World of Hyatt + American AAdvantage: partner removed takes effect.
- Jan 1, 2027. Hyatt World of Hyatt + Bilt: transfer ratio change takes effect.
Each round adds a month. The empty months on the right fill in as they are collected. Above 100, a point buys more travel than in the first month; below 100, it has been devalued.
Show the numbers as a table
| Round | Value of a point | Price in points | Price in cash |
|---|---|---|---|
| R1 · August 2026 | 100.0 | 100.0 | 100.0 |
| R2 · September 2026 | 103.4 | 92.7 | 96.3 |
| R3 · October 2026 | 92.4 | 111.5 | 103.8 |
October 2026: points bought 10.7% less than in September
Airline points fell 12.8%, while hotel points fell 5.5%. Prices in points rose 20.3% and cash prices rose 7.9%; the value of a point is the gap between the two.
| Measure | Sep | Oct | Change | Since Aug | Trips |
|---|---|---|---|---|---|
| Value of a point | 103.4 | 92.4 | −10.7% | −7.6% | 151 |
| Price in points | 92.7 | 111.5 | +20.3% | +11.5% | 152 |
| Price in cash (US$) | 96.3 | 103.8 | +7.9% | +3.8% | 164 |
| Airline points | 109.5 | 95.5 | −12.8% | −4.5% | 106 |
| Hotel points | 90.4 | 85.4 | −5.5% | −14.6% | 45 |
“Trips” counts the trips priced in both months, which are the only ones the change is based on. “Since Aug” is the change over all three months collected, each month chained onto the last. How the index is calculated.
Every release
One per month, newest first: what a point bought, which programs moved, what couldn’t be priced, and the raw file.
- Release R3October 2026 dataLatestPoints bought 10.7% less in October. Airline points fell 12.8%.Collected Oct 16 – Oct 21, 2026 · 176 searches · 151 matched to SeptemberRead the release Excel
- Release R2September 2026 dataPoints bought 3.4% more in September. Hotel points fell 9.6%.Collected Sep 4 – Sep 10, 2026 · 176 searches · 151 matched to AugustRead the release Excel
- Release R1August 2026 dataAugust sets the starting point: 159 prices across 27 programs.Collected Aug 16 – Aug 23, 2026 · 176 searchesRead the release Excel
Every way of averaging October agrees on the direction
The same 151 trips, averaged 7 ways. 7 of 7 point the same way as the index (−10.7%); the middle trip changed 0.2%.
| Point bought less | 71 | 47% |
| No change | 18 | 12% |
| Point bought more | 62 | 41% |
| 10th percentile | −41.0% | |
| 25th percentile | −14.8% | |
| Median trip | −0.2% | |
| 75th percentile | +8.5% | |
| 90th percentile | +30.3% |
A change within ±0.5% counts as no change. Percentiles: 10% of trips changed by less than the 10th percentile, and so on.
| Method | Trips | Change |
|---|---|---|
| Geometric mean (the index)Each trip counts equally; one extreme trip can’t drag it far. | 151 | −10.7% |
| Plain averagePulled hard by a few very large moves. | 151 | −2.2% |
| Median tripThe middle trip, ignoring how big the moves are. | 151 | −0.2% |
| Geometric mean, top and bottom 5% removedDrops the 7 largest rises and 7 largest falls. | 137 | −6.3% |
| Geometric mean, largest single move removedDrops the one trip whose value changed most. | 150 | −9.2% |
| Booked about 60 days ahead only | 75 | −13.3% |
| Booked about 180 days ahead only | 76 | −7.9% |
The index uses the geometric mean of each trip’s change because no single trip can dominate it and it doesn’t depend on which month comes first. Why this average.
Which programs moved the index
The index averages every trip’s change, so each program’s share of the month can be added up exactly. The three largest contributions add +1.72 log points; everything else together takes away 13.00, leaving the month’s −11.28.
A log point is 100 times the natural log of a change; for small changes it’s close to a percentage point, and unlike percentages, log points add up. For 10 of 27 programs, some trips are missing from this month’s change, because a search failed or found no award seat in one of the two months.
| Round | Searched | Priced | In the index |
|---|---|---|---|
| R1 · August 2026 | 176 | 159 90% | base |
| R2 · September 2026 | 176 | 159 90% | 151 86% |
| R3 · October 2026 | 176 | 156 89% | 151 86% |
A trip enters the change only if it was priced in both months. Why searches fail is broken down in each monthly release.
In October, 176 searches returned 156 with both a points and a cash price; 151 matched the same trip in September.
| Program | Trips | Its change | Contribution |
|---|---|---|---|
| Singapore KrisFlyer | 6 | +25.7% | +0.91 |
| IHG One Rewards | 7 | +9.9% | +0.44 |
| Atmos Rewards | 6 | +9.8% | +0.37 |
| Air India Maharaja Club | 6 | +8.7% | +0.33 |
| United MileagePlus | 6 | +8.6% | +0.33 |
| LATAM Pass | 6 | +6.6% | +0.26 |
| Korean Air SKYPASS | 4 | +3.5% | +0.09 |
| Cathay Asia Miles | 4 | +3.0% | +0.08 |
| Delta SkyMiles | 6 | +1.5% | +0.06 |
| Marriott Bonvoy | 8 | +1.1% | +0.06 |
| British Airways Club | 2 | +4.0% | +0.05 |
| Radisson Rewards | 8 | +0.1% | +0.00 |
| ANA Mileage Club | 6 | −0.9% | −0.04 |
| Air Canada Aeroplan | 6 | −1.8% | −0.07 |
| World of Hyatt | 1 | −17.1% | −0.12 |
| American AAdvantage | 6 | −4.3% | −0.18 |
| Ethiopian ShebaMiles | 4 | −6.8% | −0.19 |
| Emirates Skywards | 4 | −19.4% | −0.57 |
| Choice Privileges | 8 | −11.1% | −0.63 |
| Hilton Honors | 6 | −15.2% | −0.66 |
| China Southern Sky Pearl | 4 | −23.1% | −0.70 |
| Turkish Miles and Smiles | 6 | −16.3% | −0.71 |
| Wyndham Rewards | 7 | −15.7% | −0.79 |
| Etihad Guest | 6 | −28.9% | −1.35 |
| Virgin Atlantic Flying Club | 6 | −38.5% | −1.93 |
| Qantas Frequent Flyer | 6 | −47.8% | −2.59 |
| SAS EuroBonus | 6 | −61.1% | −3.75 |
| All programs | 151 | −10.7% | −11.28 |
Contribution = the sum of a program’s trips’ log changes ÷ all 151 trips in the index. Contributions add up to the total (−11.28). Click a program for every price and screenshot.