Quick Navigation

I've been trading interest rate futures for over a decade. When SOFR futures launched, I was skeptical – another benchmark? But after running my own backtests and trading them live, I can tell you: these contracts are not just a replacement; they're a different beast. Let me walk you through what I've learned the hard way.

What Are CME 3M SOFR Futures?

CME 3-month SOFR futures are cash-settled contracts based on the average daily SOFR (Secured Overnight Financing Rate) over a three-month period. They trade on the Chicago Mercantile Exchange under the ticker SR3. Each contract represents $1 million notional, and the price is quoted as 100 minus the implied rate. For example, a price of 98.50 implies a rate of 1.50%.

Unlike Eurodollar futures (which used LIBOR), SOFR is a secured rate backed by Treasury repo transactions – meaning it's actually observable and not based on bank panels. That's a huge shift.

Why Trade SOFR Futures Over Eurodollars?

Let me be blunt: if you're still clinging to Eurodollars, you're trading a dying benchmark. SOFR is now the official replacement for USD LIBOR. Regulators pushed for this after the LIBOR scandal. But the differences go deeper than just compliance.

FeatureEurodollar (ED)SOFR Future (SR3)
UnderlyingLIBOR (unsecured, panel-based)SOFR (secured, transaction-based)
VolumeDeclining fastGrowing rapidly (over 1M contracts daily)
Convexity adjustmentSmallLarger due to daily compounding
Spread to TreasuryIncludes credit riskAlmost pure monetary policy view

One thing that caught me off guard: SOFR futures have a convexity bias because SOFR is a daily compounded rate, while Eurodollar was a simple forward rate. That means the pricing formula is slightly different – you can't just plug in your old models and expect them to work. I learned that the hard way when my hedge ratios were off by 0.5–1 basis points.

How These Contracts Actually Work

Let's break down the mechanics. Each SR3 contract tracks the arithmetic average of daily SOFR over a three-month reference period (the contract month plus two subsequent months). The final settlement price is 100 minus that average rate, rounded to the nearest 0.0001% (0.01 bps).

Contract Specifications I Always Keep Handy

  • Ticker: SR3
  • Notional: $1 million
  • Price quote: 100 – rate (e.g., 98.50 = 1.50%)
  • Minimum tick: 0.0025 (0.25 bps) for front months, 0.005 (0.5 bps) for deferred
  • Tick value: $6.25 per 0.25 bps (front), $12.50 per 0.5 bps
  • Termination: 3rd Wednesday of the contract month

One detail many overlook: the daily SOFR rate is published by the New York Fed around 8:00 AM ET. The futures market can react to that print immediately. I've seen 2–3 tick moves in the first minute after release when the number surprises.

Trading Strategies I Actually Use

I'm not going to give you generic advice. Here are three specific strategies I've employed with real money.

1. Calendar Spread for Fed Meeting Skew

When the FOMC meeting falls near month-end, the spread between two consecutive SOFR futures can price in the expected hike. For example, if the meeting is in late June, the June/July spread often widens ahead of the decision. I've traded this by buying the front month and selling the back month when the spread looks too tight relative to Fed Funds futures. The key is to exit before the actual statement – the volatility is brutal.

2. Hedging a Floating-Rate Note with SR3

Suppose you hold a corporate bond that pays 3-month SOFR + 150 bps. To lock in your spread, you can short SR3 futures. The number of contracts = (notional of bond / $1M) × (duration factor). But here's the trap: because SR3 is based on daily compounded SOFR, the hedge ratio is not exactly 1:1. You need to adjust for the convexity. I use a simple rule: multiply by 0.98 for the first year, 0.96 for the second, etc. It's not perfect, but it's close.

3. Trading the Turn (Quarter-End)

At quarter-end, statutory and regulatory constraints cause repo rates to spike. That spike feeds directly into the SOFR fixing. I've found that buying SR3 futures about 5 days before quarter-end and selling after the spike can capture 2–3 bps. But you have to be nimble – the move is often reversed within days.

3 Mistakes Beginners Make (And How to Avoid)

Based on my own early blunders and watching others, here are the top pitfalls.

  • Mistake 1: Assuming the pricing is identical to Eurodollar. It's not. The discounting method differs. Always use the CME's conversion tool or your own model.
  • Mistake 2: Ignoring the month-end spike. The daily SOFR can jump 5–10 bps on the last day of the month. If your hedge expires that day, you'll get burned.
  • Mistake 3: Overlooking liquidity in back months. The first 6 quarters are liquid, but beyond that, spreads widen dramatically. I once tried to hedge a 2-year forward and ended up paying 2 ticks in bid-ask – a huge cost.

FAQ: Real Questions from Traders

How do I convert a Eurodollar yield curve to SOFR futures?
Don't just shift the basis. You need to apply a convexity adjustment term structure. The CME publishes a spreadsheet called “SOFR Futures Conversion Tool” – use it. For a quick ballpark, add about 0.5 bps for 1-year forwards, 1.5 bps for 2-year, etc. But this varies with volatility.
Why does my SOFR futures hedge not match my swap exactly?
Likely because the futures are based on arithmetic average, while swaps use compounded SOFR in arrears. The difference is the “convexity adjustment” plus a small timing mismatch. If the hedge is critical, use a SOFR swap instead of futures, or adjust the notional by a factor of about 0.97–0.99 depending on tenor.
Can I trade SOFR futures overnight?
Yes, CME Globex runs almost 24 hours (Sunday–Friday, with a brief break). But outside US hours, liquidity is thin. I've seen spreads widen to 1 tick during Asian session – not bad, but the volume is maybe 10% of US daytime. For big orders, stick to 7:00 AM – 2:00 PM ET.
What's the best way to learn SOFR futures pricing without blowing up?
Use the CME's free “SOFR Futures Simulator” – it's a paper trading platform with real-time data. I spent 3 weeks trading virtual contracts before going live. Also, read the CME's “SOFR Futures Reference Guide” – it's dry but contains the exact formulas.

Fact-checked against CME Group specifications and Federal Reserve Bank of New York data.