Field Notes

Your draw weight isn't what the limbs say — and it moved my FPS estimate.

My limbs are labeled 80 lb. A scale said 84.6. Here's how that 4.6 lb let us add a draw-weight term and put two builds within 0.3 fps of the chrono.

David Marcus · June 2026

Update (June 2026): a fourth chrono session replaced this post's linear model with an energy-based one — read Your bow is an energy source →

A few weeks ago I chrono'd three of my builds to see how close ArrowForge's estimated FPS came to reality. The heavier builds landed within about 2 fps, the lightest a bit further off, and every error pointed the same direction — the app was reading low. I bumped one efficiency constant, closed most of the gap, and flagged a draw-weight term as future work.

This is that follow-up. And it started with a number I'd been wrong about for months.

The limbs lied. Sort of.

My bow wears a set of limbs stamped 80 lb. That's the number I'd entered in the app, the number I'd have told you at the range, the number I trusted. So before doing any more formula work, I did the thing I should have done first: I put a draw-weight scale on the string and pulled it back to the wall.

Peak weight came up 84.6 lb — 4.6 lb over the sticker. That's not a defect. Limb ratings describe a nominal maximum, and where you actually land depends on limb-bolt position, the specific limbs, manufacturing tolerance, and how the cams time out on your bow. A few pounds either way is completely normal. The point is simply this: the printed number is a starting estimate, not a measurement.

And 4.6 lb is real energy. A heavier peak weight stores more energy in the limbs, and a bow that stores more energy throws the same arrow faster. If the app thinks I'm pulling 80 when I'm really pulling 84.6, it's working from a bow that's quietly weaker than mine. That's exactly the kind of error that reads as "the estimate is a little low across the board."

Adding a draw-weight term

The IBO speed rating manufacturers publish is measured under a fixed standard: 70 lb draw weight, 30" draw length, a 350 gr arrow, bare string. Every setup deviates from that, and the formula already corrected for draw length, arrow weight, and string accessories. It just didn't correct for draw weight — because IBO is rated at 70 lb and most of us pull more.

So I added the term the formula was missing. The industry rule of thumb is roughly 2 fps per pound of draw weight above or below the 70 lb IBO standard:

At my measured 84.6 lb, that term adds (84.6 − 70) × 2 = +29 fps before the efficiency factor. That's a big swing — which is the whole point. Pulling nearly 15 lb over the IBO standard should add real speed, and now the formula says so.

Re-tuning the efficiency factor

Adding +29 fps means the old efficiency factor (0.975) no longer fits — leave it there and every estimate jumps 8 fps high. So the factor and the draw-weight term get calibrated together against the chrono data. Fitting my two most consistent builds, the factor lands at 0.888.

That number looks alarmingly low — it reads like the bow is bleeding 11% of its energy somewhere. It isn't. The efficiency factor stopped being a pure physical "efficiency" the moment it started sharing the work with a +29 fps draw-weight term. The two only mean something together: the term credits the poundage, the factor pulls the total back down to what the chronograph actually recorded. Read either one alone and you'll draw the wrong conclusion.

The draw-weight term and the efficiency factor are two halves of one calibration. Adding speed for poundage, then scaling back to measured reality — that's the trade that keeps the output honest.

The results

Same three builds, same Garmin Xero averages, zero wind. "Before" is the old model reading my bow as 80 lb with no draw-weight term. "After" is the new model with my measured 84.6 lb and the draw-weight term.

The two heavier builds — Hornet and Bonecrusher, my most repeatable data — went from roughly −2 fps to +0.3 fps. That's about as tight as a specs-only estimate gets, and it came directly from feeding the formula a measured draw weight instead of a sticker.

The lightest build still won't behave

Freight Train, my 466 gr arrow, sat at −5.5 fps before and stayed at −5.5 fps after. The draw-weight term didn't touch it — and that's informative. If the residual were a draw-weight problem, it would have moved with the other two. It didn't, which tells me the lightest arrow's gap lives somewhere else: most likely the arrow-weight penalty (0.2 fps per grain) being a touch too steep for lighter shafts.

That's the next thread to pull. But I'm not going to guess at it — I'll chrono more builds across a wider arrow-weight range and let the data set the number, the same way it set this one.

What this means for you

If you've never put a scale on your string, the draw weight in your profile is a guess wearing a sticker's authority. A draw-weight scale is cheap, and the measurement takes thirty seconds. Enter your measured peak weight, not the limb rating — for me that one change was worth a couple of fps on every build, and it's the difference between an estimate that drifts and one that lands.

And the standing advice hasn't changed: if you own a chronograph, a measured FPS reading always beats any estimate. ArrowForge's Live Settings panel takes a chrono number directly and treats it as the source of truth. The estimate is for everyone who hasn't shot through one yet — and it's a little sharper this week than it was last week.

Honest limitations

This is still one bow, one draw length, one draw weight. The 2 fps/lb draw-weight coefficient is an industry rule of thumb, not something I've yet validated against my own data — to do that cleanly I need chrono readings at a second poundage on the same bow, so I can isolate the draw-weight effect from everything else. Until then, the term and the efficiency factor are co-calibrated on a single setup. They're honest for setups like mine; I'll keep tightening them as more data comes in.

If you've run your own chrono numbers — especially at different draw weights or with lighter arrows — I'd genuinely like to see them. More data is exactly what turns a rule of thumb into a calibrated constant.