This just in from Model Accuracy

Just one race, but the WRF does seem to compare favourably in all the test cases I have seen. The AXS should normally slightly outperform WRF I think though.

Q: Which weather source most correctly forecast the first 24 hrs of the Sydney to Hobart Race?

A: Model Accuracy knows! 

Each weather source was allowed 1 GRIB file downloaded right before the start of the race, and then compared to sailboat instrument data from a boat that actually sailed the event. Each weather file source was only allowed one weather file to ensure it was "a fair fight."

The data was analyzed for the 1st 24 hours of the event.

The rankings are as follows: 

1st Place with RMS error of 5.47 : Expedition's WRF High Res GRIB

2nd Place with RMS error of 5.80 : AXS Australian Bureau of Meteorology High Res

3rd Place with RMS Error of 7.32 : GFS Predict Wind

4th Place with RMS Error of 7.94 : EC Predict Wind

1st and 2nd place finishers did a great job of capturing the overall "trend" of the forecast which Model Accuracy takes into heavy consideration for the scoring algorithm. Their ability to correctly forecast the TWS between the 18hr-24hr mark was important to their top finish. 

Nearly all four GRIB file sources correctly forecast the TWD to be expected in the first 24 hrs, but the WRF and the AXS sources did the best job of predicting the TWS correctly. 

Did you sail the Sydney to Hobart? If so, did you use Model Accuracy to verify which GRIB file source was correctly predicting the forecast, and base subsequent routing off the top performing weather file source? Let us know if you used the software while sailing offshore!

We will be releasing the results of the real time analysis completed against GRIBs that were actually downloaded during the Sydney to Hobart race. Stay tuned for those results to follow shortly! 

Model Accuracy is still free for download until January 6th, so hurry in for a free trial of our software! 

Check out the graphs from the analysis below!