Last week, a player used the Rating Translator, entered his World Tennis Number, and got a result that was off by a full NTRP level. He's a confirmed USTA 4.0C. His UTR of 6.3 returned an NTRP equivalent of 4.1. Right on target. His WTN Singles of 28.09 returned 3.0. A full level low.

I didn't dismiss it. I didn't quietly tweak a lookup table. I dug into the data, confirmed the problem was structural, and pulled WTN from the translator entirely.

Here's what I found, why better calibration won't fix it, and what it says about how I think about accuracy at MyTennisRatings.

What I Found

The data told a clear story. WTN is not a linear scale that maps onto NTRP the way UTR does. Three structural differences make reliable conversion impossible.

Separate singles and doubles ratings. WTN gives you two numbers: one for singles, one for doubles. NTRP gives you one. A player's WTN Singles and WTN Doubles can land in completely different ranges, so any "WTN to NTRP" conversion has to pick one and ignore the other. That's a real problem when most USTA league players compete primarily in doubles.

A 40-point reverse scale with massive overlap. WTN runs from 40 (beginner) to 1 (professional). NTRP runs from 2.0 to 7.0 in half-point increments. The mapping between these two scales is not evenly distributed. Schmidt Computer Ratings found that a single NTRP level can span 10 to 30 WTN points. A "conversion" with that kind of range is not a conversion. It's a coin flip with a decimal point.

The USTA's own position. The USTA explicitly states there is no direct one-to-one comparison between NTRP and WTN. That's not a gap in their documentation. It's a deliberate statement from the governing body of the NTRP system. They built the thing, and they're telling you the two scales don't line up.

The USTA explicitly states there is no direct one-to-one comparison between NTRP and WTN. When the governing body of one system says it can't be reliably mapped to the other, you listen.

The Real Player Data

Two test cases confirmed what the structural analysis suggested.

Player A: Verified USTA 4.0C

UTR Singles: 6.3 → Translator returned NTRP 4.1 ✓

WTN Singles: 28.09 → Translator returned NTRP 3.0 ✗

His UTR and NTRP were perfectly aligned. His WTN conversion was off by a full level. That's the difference between a competitive 4.0 and someone who would be overmatched at 3.5.

Player B: Verified USTA 3.5C

UTR: correctly aligned with NTRP ✓

WTN: produced a conversion inconsistent with both NTRP and UTR ✗

Same pattern. UTR tracked. WTN didn't.

These are not edge cases. If the tool is consistently off by a full level, it doesn't matter how clean the interface is. The output is wrong. Showing it to a player as though it's reliable would be irresponsible.

Why This Can't Be Calibrated Away

My first instinct was the same one any builder would have. Adjust the lookup table. Recalibrate the ranges. Test against more data points and find a tighter mapping.

But this is not a calibration problem. The 10-to-30-point span per NTRP level means there is no mapping that works consistently. A WTN of 28 could be a 3.5 or a 4.0, depending on the player's match history, which federation reported their results, whether the number reflects singles or doubles, and how many of their matches were even captured by the ITF.

UTR works as a conversion source because its methodology is similar to the USTA's dynamic rating system. Both weight match results and score margins. Both update continuously. Both produce a single number on a scale where the relationship to NTRP is relatively stable. WTN shares none of those properties.

More data would not fix this. It would just make the inaccuracy look more precise.

The Decision

I could have left WTN in the translator with a disclaimer. "WTN conversions are approximate and may not reflect your actual level." Most tools would do that. It covers the liability without pulling the feature.

I went the other way. MyTennisRatings has a principle that matters more to me than feature count: I only publish conversions I can stand behind. That line is on the translator page now. It's not marketing copy. It's a rule I build by.

The player who reported the inaccuracy tested the system and found where it broke. That's exactly what I want users to do. The correct response to that is not to minimize the finding. It's to act on it.

I only publish conversions I can stand behind. That principle is worth more than a feature that looks good on paper but misleads players in practice.

What Stays, What Changed

The Rating Translator now supports two systems: NTRP and UTR. Those are the conversions where the data is reliable and consistent. Section strength adjustment still applies to NTRP inputs across all 17 USTA sections.

The WTN Guide stays on the site as educational content. WTN is a real system that some players have and want to understand. The guide explains how it works, how it differs from NTRP and UTR, and why the conversion problem exists. What it no longer does is pretend to convert something that can't be reliably converted.

To fill the gap WTN left, I built a tool that was already on the user feedback list: Rating Check. Enter your NTRP and UTR side by side and see if they agree. It does something more useful than a WTN conversion ever could. It shows whether your two most relevant ratings are aligned, flags gaps, and tells you if you're a potential bump candidate.

What This Means Going Forward

If you've been using MyTennisRatings since the early weeks, you've seen features get added, refined, and in this case, removed. That's how it works when a tool is built by someone who actually plays this sport. I'm not trying to ship the most features. I'm trying to ship the right ones.

If you have a WTN and you're wondering how it relates to your NTRP, the honest answer is: not reliably. The WTN Guide explains what your number means within the ITF system. The Rating Translator and Rating Check handle the conversions that actually hold up.

If you find something else that doesn't look right, tell me. That's how this gets better.

Ron Satha