Rate limit on stream with multiple #'s and accuracy of results returned


I am in the process of making an app using the Twitter API to collect some data. I wonder if anyone can help with a couple of questions about the results returned from the streaming API?

First of all I would like to know if there would be restrictions on making regular but spaced out requests for around 100 #'s or words? Essentially what I want to know is the count of tweets about certain terms or hashtags and there could be up to 100 different words I’d like the volume count of

Also I wanted to know if the 1% of tweets returned in the sample stream could be taken as an accurate representation of the number of tweets about these terms? If I only care about the volume of tweets about the term, could I take the sample results as relatively accurate in terms of the number of tweets containing that term?

Hopefully I’ve got enough across there as to what I’m asking, any help would be greatly appreciated




Sorry I just wanted to add an example to this. Let’s say I wanted to count the number of mentions each of the 92 teams from league 2 to the premiership in English football get mentioned live, would the Twitter stream…

A. Give me a relatively consistent and accurate result in terms of the number of times these teams are tweeted about, and reflect that say Man Utd would be the most tweeted term should they beat Arsenal 8-0 or something. Aaaand

B. Let me stream as many as 92 terms, always on, and pull numbers relative to the genuine volume of tweets about these terms?

Again thanks for any help you can offer