Is there any significant change in the percentage of bitcoin traffic overall that satoshidice accounts for? Increase, decrease? There was a time no too long ago that SD was the topic concerning blockchain space. Since then a lot of new services utilising bitcoin have come onto the chain. Simplistic thought: Either the utility of bitcoin has not really increased or the number of SD players has increased on par. A simple way to gauge increase utility or lack there of. Thoughts?

Good point!


curl 'https://bitcointalk.org/index.php?action=printpage;topic=80312.0' | awk -v RS='<br' '{print}' | grep -o -P '(Blockchain Tx:.*SatoshiDice Tx:.+|Results: 201[-a-zA-Z0-9]+)' | sed 's/&nbsp;//g' | less


Or graphed:


curl 'https://bitcointalk.org/index.php?action=printpage;topic=80312.0' | awk -v RS='<br' '{print}' | grep -o -P 'Blockchain Tx:.*SatoshiDice Tx:[^)]+' | grep -o -P '\d+.\d+%' |  gnuplot -p -e 'plot "-" with lines;'


Starts 2012-Jun-11, ends 2013-Apr-20
https://i.imgur.com/SZD6e9r.png

Nice work!

Now, you just need to plot the USD BTC price (at the dates corresponding to your data points) against each data point, and you'll be able to see what sort correlation exists before you go messing about with linear models.

Why would it be important to add USD BTC price to the graph? Do you mean I need to take into account the change in total number of bets placed due to increased stakes caused by BTCUSD? I also graphed the total number of bets and there is a slight hump beginning of year but now back to the usual constant slope. Sure, more $$ a play but same number of TX's. Implying that the Satoshidice TX load has not changed much. Yet despite this one would expect it's share of the blockchain to decrease. Potentially, and based on the first graph, its share is indeed decreasing, very slowly, for the past couple months. Still, this leaves me to wonder about why other forms of utility and services in bitcoin have not displaced satoshidice more.

Sorry, I thought you were implying something in your original post that you weren't - is the change in the percentage of SatoshiDice Tx related to the USDBTC price? I would expect that, since there's a minimum amount of btc that can be bet, as the exchange rate increases the minimum btc bet becomes much higher in terms of fiat and might start to price some players out of the market.

So I'd be expecting a negative correlation between number of SatoshiDice tx and USDBTC price. That recent dip in the percentage of SatoshiDice tx's in the blockchain is rather suggestive.

You could check this, not by adding the price to your chart, but by scatter-plotting the tx as a function of price - the y axis the daily change in number of tx, the x axis the daily change in price. I expect you'd need to do it on a log-log axis , or plot the logs of the daily tx change and daily price change.

I'd do it but I'm in the middle of something else right now. So ....

I don't know how to tell what wagers are real, and what are test bets, but the first bet was in block 176627 (2012-04-21 17:02:44) and a bunch more were in block 176629 (2012-04-21 17:28:41).

So I'd say we just passed the first anniversary.

This: https://blockchain.info/tx-index/4228396/1 appears to be the first ever bet.

Bets per day. Based dooglus' semi daily reports in this thread. Data was smoothed (bezier) to take into account spikes and jumps as reporting timing is not always equally delimited.

https://i.imgur.com/I1lwIiK.png

https://i.imgur.com/DYK48Fg.png