There are 4 segments of matches involving Mr. Fantoni/Mr. Nunes on YouTube from the 2013 Bermuda Bowl in Bali, including segment 2 in the semi-final against USA-1 and segment 4 in the final against Italy.

The video segments are listed at http://www.bridgescoreplus.com/data/bermuda_bowl_2013/index.html

The data for Mr. Fantoni/Mr. Nunes is at http://www.bridgescoreplus.com/data/bermuda_bowl_2013/pairs/Claudio%20Nunes-Fulvio%20Fantoni.html

The hypothesis for their opening leads from "The Video Speaks" series is: 

• The card is placed vertically when:
• the led suit contains an unseen high honor (ace, king, or queen) OR
• the led card is a side-suit singleton in a trump contract

• Otherwise, the card is placed horizontally.

Mr. Fantoni/Mr. Nunes are on lead 27 times. 5 of the leads are not on video, 1 is a diagonal lead pointing towards dummy and is ignored. Of the remaining 21 leads, 2 do not match the hypothesis. One is a horizontal lead of K♦ from KQ752, the other is a vertical lead of a trump singleton (6♦) against a sacrificial 5♦X. Co-incidentally both false positives are the last two boards on video from the final. The other 19 leads match the hypothesis (12 horizontal, 7 vertical).

The YouTube videos were originally posted by two different people. If you are aware of any more videos, please let me know. Many thanks to the crowd-source volunteers who have recently been looking at videos. There are more YouTube videos of other tournaments that remain to be crowd-sourced, if you are interested in being a volunteer please PM me and I will add you to the email list I have.

The YouTube videos contain the live stream video feed from Bali, which sometimes went down. Sometimes the match had started before the video feed was connected. This explains why some leads are missing from the video.

The final results from Bali were 1. Italy, 2. Monaco, 3. Poland, 4. USA-1.

If we assume that these videos were not viewed when creating the original hypotheses on Mr. Fantoni/Mr. Nunes, then we can apply different mathematical tests. Instead of the hypergeometric distribution ("urns in a ball") scenario, we can apply common statistical tests. I will let the professional statisticians give the numbers - there are various types of tests - and also try to explain what the results mean in language that BW readers can understand!

You are welcome to draw your own conclusions, also to consider if this is a sufficiently large set to be able to draw a conclusion. The professional statisticians can help answer both of these questions.