Barbados Traffic Analysis Challenge
'/%3e%3c/svg%3e)
Helping Barbados
$11 000 USD
Completed (~1 month ago)
Computer Vision
Prediction
Object Tracking
Video Analytics
Deep Learning
1824 joined
222 active
Start
Nov 07, 25
Close
Jan 25, 26
Reveal
Jan 26, 26
'/%3e%3c/defs%3e%3cpath fill='%23fff' d='M-120-80h240V80h-240z'/%3e%3cpath d='M-120-80h240v48h-240z'/%3e%3cpath fill='%23060' d='M-120 32h240v48h-240z'/%3e%3cg id='b'%3e%3cuse xlink:href='%23a' stroke='%23000'/%3e%3cuse xlink:href='%23a' fill='%23fff'/%3e%3c/g%3e%3cuse xlink:href='%23b' transform='scale(-1 1)'/%3e%3cpath fill='%23b00' d='M-120-24v48h101c3 8 13 24 19 24s16-16 19-24h101v-48H19C16-32 6-48 0-48s-16 16-19 24z'/%3e%3cpath id='c' d='M19 24c3-8 5-16 5-24s-2-16-5-24c-3 8-5 16-5 24s2 16 5 24'/%3e%3cuse xlink:href='%23c' transform='scale(-1 1)'/%3e%3cg fill='%23fff'%3e%3cellipse rx='4' ry='6'/%3e%3cpath id='d' d='M1 5.85s4 8 4 21-4 21-4 21z'/%3e%3cuse xlink:href='%23d' transform='scale(-1)'/%3e%3cuse xlink:href='%23d' transform='scale(-1 1)'/%3e%3cuse xlink:href='%23d' transform='scale(1 -1)'/%3e%3c/g%3e%3c/svg%3e){kind=small}
Scoring on Private and Public leaderboard
Help · 26 Jan 2026, 07:52 · 13
Hello, guys for whom have been to competition for that long, how do you under the logic behind how the scoring on private comes, because in my last competition been seen some crazy stats. Somebody with a good understanding, kindly explain better.
In this competition, the public LB consisted of ~50% of test data and the private test, the other 50%. It can of course happen that the distribution of the private set is completely different from the LB one. In such cases, you usually see quite a large shake up if CV was not done properly. The other important factor is luck. The skill+luck combination is undefeated.
Maybe luck plays a bigger role here, based on the scoring I have seen so far today. Public scoring does not give "hopeful thought" of a gurantee win
im glad you noticed, its a struggle
Maybe there is luck. But for us the highest scoring subs were also the ones that had the best CV.
Maybe just maybe
Scoring is done on in two ways: One is on the submission with the best public leaderboard score. Another is on two submissions that you select before the end of competition. Your private leaderboard score is evaluated on these two selected submissions, and the public leaderboard score is not evaluated on.
That is said, but can you check the leaderboard again and see the scoring, I think there is a way evaluation is done
what if i selected the wrong submissions that are not the best?
unfortunately we get to live with such choices. I have missed out on a podium finish a few times because of my poor sub selection.
True true, @nymfree thanks 🙏
I remember during the Agribora competition, I selected my two best scores. However, when the final scores were revealed, I was shocked to find that they performed worse than one that I had not selected and never thought would score well. This experience made me realize that sometimes, results seem to depend more on luck than skill.
indeed. but that was a forecasting competition with very little data. If the dataset was large enough, skill would have been more of a factor.
The same thing happened to me in this competition. My best score, 0.693, was not chosen because it was not the best on my public score, so I selected the one that seemed to be the best, with a private score of 0.644.