Bionan 

Thanks. I believe I created a template for the indicator and used the same template for both charts. I just attributed the difference in signal output to the nature of tick charts. I will double check these settings, because if I get a signal on one chart, I would prefer to get the same signal under the same conditions on a second identical chart. Otherwise, the results of the strategy could vary slightly due to "luck of the draw". Most of the signals seems to match up, however. I'll report what I find.

There is no "luck of the draw" with computers. When you observe different outputs using the same algorithm, it can always be attributed to different input data.
For the case you mentoined, this points to the trading hours template or the chart lookback period.

Alternatively, it is possible that you have applied the indicator to two charts with different settings. In such case you may also observe different signals.

In case that you find different signals again on two idnetical charts, I am happy to assist to find the cause.

Fat Tails 

There is no "luck of the draw" with computers. When you observe different outputs using the same algorithm, it can always be attributed to different input data. For the case you mentoined, this points to the trading hours template or the chart lookback period. Alternatively, it is possible that you have applied the indicator to two charts with different settings. In such case you may also observe different signals. In case that you find different signals again on two idnetical charts, I am happy to assist to find the cause.

Thanks. I may have a follow up question when I get back to my computer this evening (US time). I don't want to hijack this topic. What is the best way to post a followup question regarding the Auction Bars setup?

Bionan 

Thanks. I may have a follow up question when I get back to my computer this evening (US time). I don't want to hijack this topic. What is the best way to post a followup question regarding the Auction Bars setup?

Just send me a direct message.

Fat Tails 

With regard to the Auction Bars indicatoor there are two possible cases: -> You used different trading hours templates, such as a ETH, RTH or a 24/7 template for the 2 charts. In this case you will also see that the candles have a different shape. -> You used the same trading hours templates and observe exactly the same candles, but had selected different lookback periods for the two charts. The Auction Bars indicator has a comparatively long training period, as it calculates market volatility for different filters from the secondary bars added. In case that one of your charts had a comparitively small lookback period, this may have affected the generation of signals.

OK. Now I feel stupid. For accuracy, and for the sake of the indicator, I must correct my error. While the Auction Bars indicators on both charts were from the same template, the data set varied in both ways you suggested. One chart was using 30 days of historical data, and the other was using 90 days. Also, one chart was set to "CME US Futures ETH" , and the other was set to "use instrument settings". I would also imagine that these differences could have resulted in slightly different looking tick bars. Thanks for the correction, and it's nice to know that the Auction Bars indicator resulted in identical reversal signals at the same positions on two different charts.

Or, like some of us have witnessed first hand over the years, the data varies from broker feed to broker feed as well as with same broker and their own different feeds.

SunTrader 

Or, like some of us have witnessed first hand over the years, the data varies from broker feed to broker feed as well as with same broker and their own different feeds.

This also depends on the instruments charted.

For example, stocks are traded on different stock exchanges, trading venues and dark pools. This means that price discovery takes place in different locations simultaneously. Data providers cannnot connect to all those electronic exchanges and therefore will only be able to provide for an incomplete picture of all transactions.

For FOREX things are even worse, as there are many different ECNs (electronic communication networks), each of them with a different price discovery. Of course, prices do not differ a lot between those ECNs, as larger players are engaging in arbitrage. However, it is not recommended to take a data feed from FXCM and then daytrade FOREX on IDEALPRO of Interactive Brokers. The two ECNs simply have different prices.

However, for futures contracts the data should be coherent. All futures contracts are bound to a single exchange. When data comes from a single exchange, it should always be the same. At least when it is unfiltered. So I do not agree that the data always differs between broker feeds. At least unfiltered data from a single exchange should not depend on the data provider. If it is, something is wrong and you should check for the reasons.

Fat Tails 

This also depends on the instruments charted. For example, stocks are traded on different stock exchanges, trading venues and dark pools. This means that price discovery takes place in different locations simultaneously. Data providers cannnot connect to all those electronic exchanges and therefore will only be able to provide for an incomplete picture of all transactions. For FOREX things are even worse, as there are many different ECNs (electronic communication networks), each of them with a different price discovery. Of course, prices do not differ a lot between those ECNs, as larger players are engaging in arbitrage. However, it is not recommended to take a data feed from FXCM and then daytrade FOREX on IDEALPRO of Interactive Brokers. The two ECNs simply have different prices. However, for futures contracts the data should be coherent. All futures contracts are bound to a single exchange. When data comes from a single exchange, it should always be the same. At least when it is unfiltered. So I do not agree that the data always differs between broker feeds. At least unfiltered data from a single exchange should not depend on the data provider. If it is, something is wrong and you should check for the reasons.

Correct except futures charts are not always all alike.

I use TradeStation, depending on time when I log on and their trade activity load at that moment determines which of their many servers I will connect to here in FL, TX, NJ or IL. I've seen many times when the server I am connected goes down, or more likely here in South Florida we have a "loud" Summer storm pass thru messing with the electricity (of course I have UPS unit connected) causing a blip in the connection. Once I log back in charts all do a dance with some being slightly different. Especially tick charts although I'm not much of a fan of those anymore.

SunTrader 

Correct except futures charts are not always all alike. I use TradeStation, depending on time when I log on and their trade activity load at that moment determines which of their many servers I will connect to here in FL, TX, NJ or IL. I've seen many times when the server I am connected goes down, or more likely here in South Florida we have a "loud" Summer storm pass thru messing with the electricity (of course I have UPS unit connected) causing a blip in the connection. Once I log back in charts all do a dance with some being slightly different. Especially tick charts although I'm not much of a fan of those anymore.

Reloading data should fix it?

Fat Tails 

Reloading data should fix it?

Yes it should. Yet many times nah ah.