Built by Grok 4.6 HighReports Wednesday, October 21, 2026. Prepared Thursday, September 24, 2026. This is the Q3 2026 print (quarter ended September 30). It has not been reported. Written highlights at 6:00 a.m. Central, conference call at 7:30 a.m. Central. Last reported quarter is Q2 2026, released the morning of July 22, 2026. CME is a calendar-year company. The homepage calendar has the date right and the period label wrong: it tags October 21 as 2026Q4. That label is one quarter ahead. This page is 2026Q3.
CME | Earnings Preview
| Q3 revenue / EPS / ADV | Not guided · the bar is FMP, not a company range | FY2026 adj. opex ex-license | $1.695B · unchanged since February · conservative |
| FY2026 capex / tax | About $85M · tax 23.5–24.5%, written to the low end | Known Q3 volume | July ADV +23% · August +6% · September not out |
CME reports Q3 before the open on Wednesday, October 21. The quarter is not in the numbers yet. Daloopa's latest fiscal quarter is still 2026Q2. On July 22 the company printed revenue of $1,706.2 million, +0.8% versus $1,692.0 million, and adjusted diluted EPS of $2.99, +1.0% versus $2.96. That was a hard deceleration from Q1, when revenue grew +14.5% ($1,880.1 million versus $1,642.3 million) and adjusted EPS grew +20.0% ($3.36 versus $2.80). The slowdown is 1,370 bps on the revenue growth rate and 1,900 bps on the EPS growth rate. GAAP operating margin compressed 180 bps, from 66.7% to 64.9%. Clearing and transaction fees fell 2.6%. Market data rose 20.2%.
The street is asking for a reacceleration off an easy quarter, not a continuation of Q2. FMP, pulled 2026-09-24, has Q3 revenue of $1,692.3 million and EPS of $2.94 (average $2.93579). Versus Q3 2025 that is +10.1% on $1,537.6 million and +9.5% on $2.68. Q3 2025 itself was −3.0% on revenue versus $1,584.4 million and flat on adjusted EPS versus $2.68. ADV that quarter was 25,322 thousand contracts, −10.5%. A +10% revenue print would be a rebound off a down quarter. It would still be about 4 points slower than Q1 and about 9 points faster than Q2.
July did not survive August. On the Q2 call, July-to-date volume was tracking 18% ahead of last year. July finished at a record 27 million contracts, +23%. August was 29.7 million, still the second-highest August, but only +6%. Rates went from +17% to +3%. Equity index went from +48% to +7%. Energy went from +9% to +3%. A simple average of the two months is 28.35 million, about +12% versus the Q3 2025 ADV base. It is not day-weighted, and September is not in it. The September 16 hike, to 3.75–4.00% on a 12–0 vote, sits in the unpublished month. Consensus at +10% revenue is closer to "August, plus market data, plus the 1% to 1.5% fee change" than to "July annualized."
Rate per contract is why volume is not revenue. Q3 2025 RPC was $0.702, the high mark of the recent year. Q2 2026 was $0.678, −1.7% versus the year-ago quarter $0.690, even after the April 1 fee increase. The Q3 hurdle is that Q3 2025 rate. It is not a Q2-versus-Q3 growth rate. Micros were 54% of July equity-index ADV and 51% of August equity-index ADV. Holding Q2's rate on this quarter's volume means clearing-fee growth runs a few points behind ADV. Market data is the offset that does not need the Fed. It has accelerated for four straight comparable quarters: +13.6%, +14.5%, +15.2%, +20.2%, to $238.1 million. About a fifth of that Q2 step was audit and catch-up payments that management said will not annualize cleanly.
The only numbers they guide are costs, and those are conservative. Adjusted operating expense excluding license fees is still $1.695 billion, capex about $85 million, and the tax rate 23.5% to 24.5%, with the written commentary at the low end. H1 ex-license expense was $817 million ($511.7 million minus $106.8 million, plus $521.2 million minus $109.1 million), 48% of the year, so the guide only gets used if Dallas occupancy and cloud spend fill the second half. License fees sit outside that dollar and grew +13.4% in Q2. There is no tariff cost line in this setup. The policy item inside the quarter is the rate hike, which shows up as hedging volume, not as a goods-tariff bridge.
What to watch on October 21
- Whether September ADV looks like July's +23%, August's +6%, or something the hike pulls back toward the Treasury tape (August Treasuries were +15% even while headline rates were +3%).
- Where RPC lands versus Q3 2025's $0.702 and Q2's $0.678. Micro mix is the pressure valve.
- Whether market data stays near +20% or fades toward the mid-teens once the $7.4 million of Q2 catch-up payments drop out.
- A reaffirmation of $1.695 billion and the low end of the tax band. A cut would be the surprise. They have not previewed one.
- Perpetual futures and the CFTC case. That is a franchise argument. It is not a Q3 fee line.
- IBKR after the close on October 15 is the only company-confirmed similar-business print before CME. ICE and Tradeweb are October 29. Nasdaq's October 20 date is still an estimate.
The EPS streak is real and small. On FMP, adjusted EPS beat in 11 of the last 12 quarters and in all of the last four. The last four surprises were +1.9%, +0.7%, +0.6%, and +2.7%. A large automatic beat is not what that record supports. Q1 2026 is the one quarter print-day wires called a miss against a higher bar than FMP retained. Internal models, Visible Alpha, and Bloomberg were not available.
CME does not guide quarterly or full-year revenue, adjusted EPS, margin, ADV, or rate per contract. There is no range to beat on the income statement. The October 21 bar is FMP. The framework still in force was set on the February 4 call, left unchanged in April, and restated in the Q2 written commentary. The spoken July 22 script did not restate it and did not withdraw it.
Classification: not an aggressive guide, and not a classic sandbag, because the income statement is not guided. On the expense dollar they do set, the pattern is conservative. In 2025 they opened at $1.650 billion, cut twice, and landed on $1.625 billion. The 2026 figure of $1.695 billion is +4.3% versus that outturn and +3.7% versus the guide that was live at this point last year, $1.635 billion. It has not been trimmed. H1 used 48% of it. The residual $878 million for the second half is the Dallas and cloud step-up Fitzpatrick already described, not unused slack she has offered to give back.
Tax is conservative versus the midpoint and not a free pass through the floor. The band was reset 100 bps higher than the 22.5–23.5% range that 2025 finished above. Q2's adjusted rate of 22.3% is already 120 bps under the new floor. The commentary's "low end" instruction is not to annualize 22.3%. Capex of about $85 million is slightly tight if the second half resembles 2025, and it is not an EPS driver. The 1% to 1.5% fee comment is a rate bridge on activity similar to 2025. Activity is not similar to 2025. Applied to Q3 2025 revenue it is only about $1.55–$1.56 billion, roughly 8% below consensus, which is the wrong benchmark.
| Live guide | Into Q3 2025 | Into this print | Read |
|---|---|---|---|
| Adj. opex ex-license | $1.635B | $1.695B | Conservative. +4.3% vs the 2025 landing. H1 is 48% of the dollar. Not cut in 2026. |
| Capex | $90M | $85M | Slightly tight, low materiality. H1 cash purchases of property were $45.4M, 53% of the guide. |
| Adj. effective tax rate | 22.5%–23.5% | 23.5%–24.5% | Range +100 bps. Commentary at the low end. Q2 printed 22.3%, under the floor. |
| Q3 revenue / EPS | Not guided | Not guided | FMP $1,692.3M and $2.94. +10.1% and +9.5% vs Q3 2025. In line to slightly conservative versus July–August volume, before RPC. |
FMP has no field that maps to ex-license operating expense, capex, or the adjusted tax rate. Its Q3 SG&A average of about $44 million is not CME's operating-expense line and is not differenced against $1.695 billion. FMP EBITDA of $1,377.5 million on the Q3 revenue estimate is about an 81% margin. Q3 2025 adjusted operating income was $1,051.0 million on $1,537.6 million, the 68% margin the call described. Those are different constructs.
| FMP consensus | Revenue | Rev YoY | EPS | EPS YoY | What the comp is |
|---|---|---|---|---|---|
| Q3 2026e | $1,692.3M | +10.1% | $2.94 | +9.5% | Vs $1,537.6M and $2.68. Easy revenue quarter, flat EPS quarter. 7 / 10 analysts. |
| Q4 2026e | $1,736.5M | +5.3% | $2.96 | +6.9% | Vs $1,648.7M and $2.77. Slower, because Q4 2025 was already +8.1%. |
| FY2026e | $7,017.8M | +7.6% | $12.25 | +9.3% | Vs the 2025 quarterly sum of $6,520.6M and $11.21. H1 revenue was already +7.6%. |
| FY2027e | $7,376.4M | +5.1% | $12.92 | +5.5% | Not guided. YoY is versus the FY2026 FMP averages, not versus a company plan. |
Q3 2026 has not printed. Guide columns are company guidance for that period. "Not guided" means neither the July call nor the standing February framework gave a quarterly number. Consensus exists only at consolidated revenue and EPS. Internal is n/a on every row. Percent differences on revenue and EPS are versus the year-ago quarter, because there is no guide midpoint.
Current quarter — Q3 2026
| Metric | Guide | Consensus | Internal | Vs the comp that matters |
|---|---|---|---|---|
| Total revenue | Not guided | $1,692.3M | n/a | +10.1% vs $1,537.6M. Q2 was only +0.8%. |
| Adj. diluted EPS | Not guided | $2.94 | n/a | +9.5% vs $2.68, which was 0.0% YoY. |
| Clearing and transaction fees | Not guided | n/a | n/a | Q3 2025 was $1,227.9M, −5.3% YoY. Q2 2026 was $1,352.5M, −2.6%. |
| Market data | Not guided | n/a | n/a | Base $202.5M, +13.6%. A repeat of Q2's +20.2% is about $243M, roughly a quarter of the consensus revenue increase. Arithmetic, not a forecast. |
| Total ADV | Not guided | n/a | n/a | Q3 2025 25,322 thousand, −10.5%. July +23%, August +6%. |
| Average rate per contract | Not guided | n/a | n/a | Hurdle is $0.702, not Q1's $0.652. Q2 was $0.678. |
| GAAP operating margin | Not guided | n/a | n/a | Q3 2025 63.3%. Q2 2026 64.9%, −180 bps YoY. Do not mix with the 69.5% adjusted margin on the call. |
The line that decides whether +10% is earned is mix, not a guided dollar. Q2 rates revenue was $433.0 million, −6.2%. Equity revenue was $324.0 million, +7.4%, lagging equity ADV of +12.7% because equity RPC fell. Energy revenue was $187.0 million, −14.0%, after +21.9% in Q1. Metals revenue was $76.7 million, −9.9%, on flat ADV. Equities are the only major complex that was still accelerating on a reported-quarter basis. Energy is the drag. Q3 2025 was down in rates, equities, FX, and energy, so the comp is broad, not one product.
Eight reported quarters — trajectory into the print
Same-quarter YoY only. Q3 2026 is not a column. Q2 total ADV below is the product-line total of 29,843 thousand. The KPI the call rounded to 29.8 million is 29,800 thousand, 43 thousand lower. The −1.2% uses the product-line pair.
| Metric | Q3 2024 | Q4 2024 | Q1 2025 | Q2 2025 | Q3 2025 | Q4 2025 | Q1 2026 | Q2 2026 |
|---|---|---|---|---|---|---|---|---|
| Revenue ($M) | 1,584.4 | 1,525.3 | 1,642.3 | 1,692.0 | 1,537.6 | 1,648.7 | 1,880.1 | 1,706.2 |
| Revenue YoY | +18.4% | +6.0% | +10.4% | +10.4% | −3.0% | +8.1% | +14.5% | +0.8% |
| Adj. diluted EPS | $2.68 | $2.52 | $2.80 | $2.96 | $2.68 | $2.77 | $3.36 | $2.99 |
| Adj. EPS YoY | +19.1% | +6.3% | +12.0% | +15.6% | 0.0% | +9.9% | +20.0% | +1.0% |
| Market data ($M) | 178.2 | 181.6 | 194.5 | 198.1 | 202.5 | 208.0 | 224.1 | 238.1 |
| Market data YoY | — | — | — | — | +13.6% | +14.5% | +15.2% | +20.2% |
| Total ADV (000) | 28,289 | 25,503 | 29,768 | 30,217 | 25,322 | 27,353 | 36,231 | 29,843 |
| Total ADV YoY | — | — | — | — | −10.5% | +7.3% | +21.7% | −1.2% |
| RPC ($) | 0.666 | — | — | 0.690 | 0.702 | — | 0.652 | 0.678 |
| GAAP op. margin | 64.6% | 62.1% | 67.5% | 66.7% | 63.3% | 61.8% | 69.7% | 64.9% |
| GAAP margin YoY | — | — | — | — | −130 bps | −30 bps | +220 bps | −180 bps |
Revenue growth went +14.5% to +0.8%. Adjusted EPS went +20.0% to +1.0%. Margin expanded in the volume spike and compressed when volume stalled. Market data is the only line that has accelerated through that stall. Q4 2024, Q1 2025, and Q4 2025 RPC were not in the series used for this table, so those cells are blank rather than filled from a second definition.
Asset-class ADV, thousand contracts
| Metric | Q3 2024 | Q4 2024 | Q1 2025 | Q2 2025 | Q3 2025 | Q4 2025 | Q1 2026 | Q2 2026 |
|---|---|---|---|---|---|---|---|---|
| Interest rates | 14,881 | 13,244 | 15,029 | 15,472 | 13,378 | 13,010 | 18,674 | 14,532 |
| Rates YoY | — | — | — | — | −10.1% | −1.8% | +24.3% | −6.1% |
| Equities | 7,407 | 6,343 | 7,997 | 7,661 | 6,278 | 7,738 | 8,655 | 8,633 |
| Equities YoY | — | — | — | — | −15.2% | +22.0% | +8.2% | +12.7% |
| Energy | 2,571 | 2,519 | 2,903 | 3,082 | 2,295 | 2,523 | 3,985 | 2,667 |
| Energy YoY | — | — | — | — | −10.7% | +0.2% | +37.3% | −13.5% |
| Metals | 728 | 673 | 732 | 943 | 825 | 1,441 | 1,682 | 941 |
| Metals YoY | — | — | — | — | +13.3% | +114% | +130% | −0.2% |
| Agriculture | 1,614 | 1,755 | 1,958 | 1,964 | 1,712 | 1,787 | 2,042 | 2,080 |
| Agriculture YoY | — | — | — | — | +6.1% | +1.8% | +4.3% | +5.9% |
| FX | 1,089 | 969 | 1,149 | 1,096 | 834 | 853 | 1,193 | 989 |
| FX YoY | — | — | — | — | −23.4% | −12.0% | +3.8% | −9.8% |
Rates revenue YoY went from +19.5% in Q1 to −6.2% in Q2, a 2,570 bp deceleration, because the ADV spike did not repeat and tiering had already cut rates RPC to $0.457 in Q1 before it recovered to $0.480. Metals ADV went from +130% to −0.2%, and metals revenue was already lagging the contract boom (Q1 revenue only +67%). Agriculture is a steady mid-single-digit grower and does not decide a $1.69 billion print. YoY cells before Q3 2025 are blank because the prior-year quarter is outside this window.
FQ+1 — Q4 2026
No quarterly guide. The February items still cover the year that includes Q4. They are not a Q4 range. Treasury Link is a Q4 2026 launch, pending review. That is a product date, not a revenue guide. Securities clearing is scheduled for December 7, after this print and inside Q4, ahead of the SEC cash-Treasury mandate on December 31. It does not change Q3 fees.
| Metric | Guide | Q4 2025 actual | FMP consensus | Read |
|---|---|---|---|---|
| Revenue | Not guided | $1,648.7M | $1,736.5M | +5.3% YoY. Growth slows from the +10.1% Q3 bar because Q4 2025 was already +8.1%. |
| Adj. diluted EPS | Not guided | $2.77 | $2.96 | +6.9% YoY. EPS range in the file is $2.89–$3.05. |
| ADV / RPC / margin | Not guided | ADV 27,353 thousand | n/a | Nothing on the July call previews a Q4 volume or margin figure. |
H1 2026 revenue was $1,880.1 million plus $1,706.2 million, or $3,586.3 million. FY2026 consensus of $7,017.8 million leaves about $3.43 billion for the second half. FMP's Q3 plus Q4 sum is about $3.43 billion. The street has a normal step-up after Q3, not a second-half air pocket and not another Q1.
FY framework still in force — and FY2027, which is not guided
| Metric | FY2023 | FY2024 | FY2025 | FY2026 guide | Score |
|---|---|---|---|---|---|
| Opex ex-license, opening | $1.490B | $1.585B | $1.650B | $1.695B | 2023 overspent and was never cut. 2024 in line. 2025 cut twice and landed on the cut. 2026 not cut. |
| Opex ex-license, actual | $1,525.5M | $1,589.6M | $1,624.8M | H1 $817M, 48% | 2025 actual is $1,995.8M minus $371.0M. 2023 and 2024 are the same subtraction in the guidance note. |
| Ex-license YoY | — | +4.2% | +2.2% | Guide +4.3% | Q1 2026 ex-license growth was +7.1% and slowed to +4.3% in Q2, back on the full-year rate. |
| Capex guide | $100M | $85M | $90M | $85M | Cash purchases missed the guide by more than 10% in 2024 and beat it by 7% in 2025. Do not build an EPS view off $85M. |
| Revenue / EPS | Not guided | Not guided | Not guided | Not guided | FMP FY2026 $7,017.8M and $12.25. FY2027 $7,376.4M and $12.92, with no company guide at all. |
License fees are the caveat on the expense beat. Q2 license fees were $109.1 million, +13.4% versus $96.2 million. Total adjusted operating expense was $521.2 million, +6.1% versus $491.2 million, about 180 bps faster than ex-license growth. Hitting the ex-license guide can still leave total expense growth above 4%.
Management enters October 21 with the cost stack unchanged and the activity story already rewritten once. Confidence in the franchise is high. Confidence that July's run-rate survives into a +10% revenue quarter is not something they have claimed. They have been right, twice, that an intra-quarter volume comment describes the month they are in. They have not claimed it describes the quarter.
The useful contradiction is on perpetuals, not on expenses. Duffy's Q2 message was that commercial customers do not want the product, that the contracts are swaps under Dodd-Frank, and that CME's crypto notional still dwarfs the Kalshi bitcoin perpetual. The street's Hold-heavy mix (FMP: Buy 14 / Hold 18 / Sell 5) is an argument about the franchise and the multiple, not about a Q3 fee hole. Kalshi's September stock-perp and margin filings landed after most of the quarter's volume was done. They will take airtime. They should not be added to the $2.94. Equity perps are not approved.
The second gap is July versus the model. Management put +18% July-to-date on the table on July 22. The street's +10% revenue and +9.5% EPS estimate, set with July and August already public, did not chase it. That skepticism fits August and the RPC mix. It leaves September as the only unpublished swing.
| Call | How they framed it | What they did to the guide | Tone |
|---|---|---|---|
| Q3 2025 · Oct 22 | Second-highest Q3 ADV, but revenue down 3% from a strong 2024. Market data over $200M for the first time. | Cut ex-license expense to $1.625B. Other guidance unchanged. The cut was lower cloud spend, not a demand warning. | Cost-disciplined, volume-defensive. |
| Q4 2025 · Feb 4 | Record year. Revenue +6%, adjusted EPS +9% on the call. Fee package worth 1% to 1.5% only if activity looks like 2025. | Set 2026: $1.695B, capex $85M, tax 23.5%–24.5%. Reinvestment for 24/7 crypto, securities clearing, and event contracts sits inside the dollar. | Confident, and spending again. |
| Q1 2026 · Apr 22 | Record ADV in all six asset classes. RPC decline explained as tiering doing its job. April already a little softer. | Affirmed $1.695B. Comfortable, while flagging higher variable cost, occupancy, and cloud still ahead. Did not raise the number after the strongest quarter. | Operationally euphoric, explicitly unwilling to extrapolate Q1. |
| Q2 2026 · Jul 22 | Record Q2 revenue, ADV within 1% of last year, market data +20%. Then a long argument that the quarter had been overshadowed by perps. July-to-date +18%. | Spoken script did not restate the three figures. Written commentary reaffirmed $1.695B and $85M and moved tax to the low end of the unchanged band. | Strategically combative, financially steady. The Q1 celebration is gone. The cost guide did not move. |
Direction of the language: from "we are cutting the guide" (Q3 2025), to "here is the new year" (Q4 2025), to "comfortable, but the second half costs more and April is softer" (Q1 2026), to "volumes are ahead in July, tax at the low end, and we are not re-trading the expense number out loud" (Q2 2026). No withdrawn guide. No delayed release. The September 4 notice confirmed October 21 before the open. The CFO seat is still Fitzpatrick's for this print. Jack Tobin, named September 9, becomes deputy CFO in November and CFO in March 2027, when Fitzpatrick becomes CEO. That is an orderly bench move, not a pre-earnings exit. It does mean this call is one of the last times she defends $1.695 billion as CFO.
What would change the guide, versus what would just change the quarter: a reaffirmation is the base case. A cut to the expense dollar would mean Dallas, cloud, and variable costs are not filling the $878 million second-half allowance. A cut to the tax floor would mean they are willing to bank Q2's 22.3%, which the written commentary declined to do. Neither has been previewed. The number that can miss without any guidance change is revenue.
transcripts/CME_FY2026Q2.txt, CME_FY2026Q1.txt, CME_FY2025Q4.txt, and CME_FY2025Q3.txt in the run folder. The Q2 written commentary is the Daloopa document that still carries the three guide tags. No sell-side conference after July 22 revised them. Piper Sandler on June 4 predates the last restatement.Consensus does not publish ADV, RPC, or asset-class revenue. What it is underwriting is the total: $1.692 billion and $2.94. That fits an easy Q3 2025 base and a cooled August tape plus market data. It does not fit a full-quarter repeat of July. CME is the listed-derivatives leader. The quarter is still volume times rate, plus a market-data annuity that does not depend on the day's volatility.
| Catalyst | Last print | Already known in Q3 | Why it matters for $2.94 |
|---|---|---|---|
| Rates and the September hike | ADV 14,532 thousand, −6.1%. Revenue $433.0M, −6.2%. RPC $0.480, flat vs $0.481. | July rates ADV 12.6M, +17%. August 16.7M, only +3%, but Treasuries +15% to 11.8M. FOMC hiked 25 bps on Sept 16 to 3.75–4.00%, 12–0. The Oct 27–28 meeting is after the print. Minutes are due Oct 7. | Swing factor. The +10% total does not work if rates stay a mid-single-digit revenue drag the way they were in Q2 and in Q3 2025 (−6.4%). September has to look more like the Treasury tape than like July's headline. |
| Equities, micros, single-stock futures | ADV 8,633 thousand, +12.7%. Revenue $324.0M, +7.4%. RPC $0.605 vs $0.635, −4.7%. | Single-stock futures launched July 27. July equity ADV 8.2M, +48%, led by Micro Nasdaq-100 +159%. August 6.8M, +7%. Micros still about half of equity-index ADV. No single-stock ADV breakout. | Volume positive, rate negative. Revenue has lagged ADV by about 5 points. A partial quarter of single-stock futures is a 2027 retail story, not the reason the estimate is $1.69B. |
| Market data | $238.1M, +20.2% vs $198.1M. 33rd straight up quarter on the call. About $7.4M of audit and catch-up, vs $3.8M in Q1. | No intra-quarter disclosure. The January 1 price change is fully in the run-rate. Nothing in the news flow points to a subscriber reversal. | The part of consensus least dependent on September. A fade back to the mid-teens leaves volume to do more of the work. Do not roll the catch-up payments forward. |
| Energy, and metals as a side bet | Energy ADV 2,667 thousand, −13.5%. Revenue $187.0M, −14.0%. Metals ADV flat, revenue −9.9%. | July energy +9%, August +3%. August metals about 1.0M, +48%, with micros heavy. 1-ounce gold went 24/7 in late July. The CFTC stayed 24/7 WTI on July 9. The August 30 oil-hours launch did not happen. | Energy is a drag risk and is large enough to matter. August's +3% does not repair a −14% revenue comp. Metals can reaccelerate on an easy $79.4M base, but revenue follows RPC, not the +48% headline. |
| Perps versus CME crypto | Crypto ADV 250 thousand, inside equities. The Q2 call said the suite grew 32% versus Q2 2025. Q2 2025 is not in this quarterly series, so that YoY is call color, not a second tag. Q3 2025's 340 thousand is a different quarter and is not the growth base. | July 237,000 contracts. August 175,000. Both sit under the Q3 2025 quarterly record. That is a different quarter, not a year-ago month. 24/7 crypto has been live since May 29, so this is the first full quarter. Kalshi filings in September are not approved. | Narrative risk, not the revenue bridge. Crypto is a few percent of equity contracts and a smaller share of fees. Consensus EPS does not embed a crypto-fee hole. |
| Expense, tax, and the fee tweak | Guide unchanged: $1.695B ex-license, tax 23.5%–24.5%. GAAP margin −180 bps. | No guide change since February. Q2 tax already under the floor. License fees growing faster than the guided bucket. Variable expense moves with the same activity that helps revenue. | Stable. EPS +9.5% versus revenue +10.1% says the street expects essentially no operating surprise. The risk is mix-driven margin, not a cost-guide cut. A walk-back of the 1–1.5% fee benefit would be new. |
Better than $1.692 billion and $2.94: September rates hold August's 16.7 million with the Treasury bid intact, equity RPC stops falling, market data grows another 20% on the $202.5 million base, and August metals rather than July metals are the quarter. That is how a company that has beaten the last four EPS prints by a few cents beats again. It does not require perps to go away.
Worse: September cools further from +6% once the hike is digested, energy stays on a −14% revenue run-rate, metals revenue falls while contracts look fine, and market data fades from +20% toward the +14% trend that held through Q4 2025. A guide cut is not the base case.
Event contracts are strategic and not this print's number. The Q2 call cited about 525 million contracts since launch, ADV over 4 million, and no clean rate capture. Agriculture at +5.9% ADV is too small to move the total. International had the same shape as the complex: Q2 non-U.S. ADV 9.1 million was −1.1%, July was +32% to 8.8 million, and August was +10% to 8.6 million.
The window is July 23 through September 24. The Q2 release, the July 22 call, and the September 4 earnings-date notice are the setup, not the news. Nothing in the window replaces $1.695 billion, $85 million, or 23.5–24.5%. September ADV has not been released. Two of the three months are known, and the growth rate decelerated without turning negative.
| Date | Item | Source | Read for this print |
|---|---|---|---|
| Aug 4 | July ADV a record 27 million, +23% | CME Group / PR Newswire | Ahead of the call's +18% July-to-date comment. Rates +17%, equities +48%, energy +9%, international +32%. Micro E-mini Nasdaq-100 ADV +159% to 3.0 million. This is the strong month. The micro share is the RPC caveat. |
| Sep 2 | August ADV 29.7 million, +6%, second-highest August | CME Group / PR Newswire | The growth rate cooled by roughly 17 points. Rates +3% with Treasuries +15%. Equities +7%. Energy +3%. Metals +48%. International slowed from +32% to +10%. Still above the Q3 2025 ADV base. No longer an 18–23% quarter. |
| Sep 16 | FOMC hikes 25 bps to 3.75–4.00% | Federal Reserve statement; Reuters, Sept 16 | 12–0, the first hike since July 2023, with the dots pointing to another hike in 2026. It can help late-September rates. It cannot rewrite July or August. It is also the variable-expense catalyst Fitzpatrick tied to activity. |
| Sep 2 | CFTC moves to dismiss the perpetual-futures suit | D.D.C. case 1:26-cv-02157; The Block | Unresolved into the print. The dispute is whether bitcoin perpetuals are futures or swaps. Competitive overhang versus Kalshi and Coinbase. Not a Q3 fee item. |
| Sep 9 | Jack Tobin named next CFO | CME Group / PR Newswire | Deputy CFO in November, CFO in March 2027, when Fitzpatrick becomes CEO. John Marchese named general counsel. Internal succession. Fitzpatrick still owns this print and the expense guide. Not a guidance red flag. |
| Sep 10 | CME Securities Clearing dated Dec 7 | CME Group / PR Newswire | SEC-registered clearing for Treasury cash and repo, pending approvals, ahead of the Dec 31 cash mandate and the June 30, 2027 repo mandate. Strategic for rates. Post-print. The spend was already inside the $1.695B. |
| Jul 24 – Sep 14 | 24/7 one-ounce gold, then 100-ounce silver | CME Group | Weekend gold passed 140,000 contracts and $600M notional by mid-September. First silver weekend was 2,387 contracts. Confirms the retail metals bid behind August's +48%. Notional is not a large fee pool yet versus the benchmark contracts. |
| Jul 27 / Aug 11 / Sep 22 | Single-stock futures live; compute futures Oct 5; BCH and UNI futures Oct 19 | CME Group; Bloomberg on the equity launch | Franchise breadth. Single-stock futures had only a few July sessions inside the July ADV print and no disclosed ADV. Compute, Bitcoin Cash, and Uniswap launch after quarter-end. Talking points for the call, not Q3 fee lines. H1 crypto ADV of 279,800 did not hold: August was 175,000. |
| Aug 6 | Regular dividend held at $1.30 | CME Group / PR Newswire | Same rate as the June regular dividend. Continuity, not a raise. Does not change the operating setup. |
The operating read-through is a fast start that cooled. The non-volume news is franchise building — securities clearing ahead of the Treasury mandate, 24/7 metals, compute futures — plus an internal CFO handoff and an unresolved fight over who gets to list crypto perpetuals. Open-interest records in Henry Hub (September 1) and FX futures (September 4) are positioning, not a stated ADV acceleration. August energy ADV was only +3%, and August FX ADV was 730,000 without a published category growth rate.
On the FMP series, CME is a consistent adjusted-EPS beater and a mixed, small revenue surprise. 11 beats, 0 misses, and 1 exact in-line out of the last 12 reported quarters (91.7%), and 4 of 4 over the last four (Q3 2025 through Q2 2026). Revenue is 8 of 12 (66.7%) and 4 of 4 recently. EPS beat size deteriorated from an average +2.1% in the oldest four quarters to +1.2% in the middle four, then stabilized rather than kept falling: the last four average +1.5%, and that average is carried by Q2 2026 at +2.7%. The three prints before it averaged about +1.1%. This is not a consistent misser. It is a volume-transparent franchise whose residual surprise has shrunk to a few cents.
FMP's reported EPS field matches the Daloopa adjusted diluted EPS series in all 12 quarters. The comparisons below are adjusted, not GAAP. 2025 Q1 EPS was an exact match at $2.80 versus $2.80 and is in-line, not a beat. Q3 2026 is not on the map.
The one conflict is Q1 2026. FMP retains EPS of $3.34 and revenue of $1,875 million, a small beat on both against $3.36 and $1,880.1 million. Print-day Zacks, Benzinga, Seeking Alpha, and StockStory called adjusted EPS a miss versus $3.37–$3.38 and revenue a miss versus about $1.89–$1.91 billion. That one-cent gap flips Q1 from a beat to a miss. It is footnoted, not substituted. Q2 2026 was a beat on both FMP and the print-day wires, $2.99 versus $2.91. The $3.01 / $1.74 billion figure in an older workspace was a pre-print aggregate and was not the print-day consensus.
| Surprise vs FMP | Q3 23 | Q4 23 | Q1 24 | Q2 24 | Q3 24 | Q4 24 | Q1 25 | Q2 25 | Q3 25 | Q4 25 | Q1 26 | Q2 26 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Adj. EPS | +1.4% | +3.9% | +2.0% | +1.2% | +0.8% | +2.4% | 0.0% | +1.7% | +1.9% | +0.7% | +0.6% | +2.7% |
| Revenue | −0.3% | +0.8% | +0.8% | −0.1% | −0.3% | +0.7% | −0.2% | +0.4% | +0.5% | +0.1% | +0.3% | +1.4% |
Green is a beat, using #1E8449. Solid green is a surprise of at least 2.5%. Yellow is the Q1 2025 EPS tie. Red text is a miss, and every revenue miss in this window is inside 0.3%. No EPS cell is red on the FMP series. The Q1 2026 EPS cell is FMP's +0.6%. Print-day Zacks was about −0.4%. Read left to right: the EPS beats got smaller, then Q2 re-widened them without restoring the 2023 spread. Revenue never left a 2% band.
The mechanism is rate per contract and market data, not volume. CME publishes monthly ADV, so the top line is mostly known. On the Q1 call Duffy said the rate is hard to figure out as the client base grows, and Fitzpatrick tied $0.652 to tiering that still produced a large clearing-fee increase because ADV was a record. On the Q2 call the rate was $0.678, −1.7% versus Q2 2025's $0.690, market data did not slow, and volume was within 1% of the year-ago record. That is why the beat re-widened on a flat volume comp. 2025 Q3 is the other template: revenue down 3%, RPC at $0.702, market data over $200 million for the first time, and a +$0.05 EPS beat versus the FMP estimate of $2.63 even though year-over-year EPS growth was zero.
Read-through for October 21. The bar is modest in surprise terms and not modest in year-over-year terms: about +10% on both lines versus a quarter that was down on revenue and flat on EPS. If September looks like August, the estimate is about right and the bull case is market-data acceleration, not a volume surprise. If September reaccelerates with the hike, $2.94 is light versus the last four beats of $0.05, $0.02, $0.02, and $0.08, and the beat still usually lands inside 3% on EPS. A miss would come from September volume or rate per contract, not from a management forecast that was set too high. There is no revenue or EPS guide to beat.