As you know setting a 'percentage expectation' for disruptive events is common in any operating plan, QoS plan, security plan etc. Contingency (forward planning) of something that might/may happen as a percentage factored in advance in order to ensure support being available is standard practice.
The author Henry Basset's 'Red Sky Alliance' records in his blog (http://henrybasset.blogspot.co.uk/2013/02/attackers-collaborate-defenders-are.html):
"20-50 compromised computers per day (7 days/week) could (should) be expected."
The article does not state that 20-50 will happen daily. Moreover, it is very difficult to reconcile how events happening on weekdays will equally occur in the same percentage on weekends due to staff not working at that time, thus terminals not being used in the workplace, which is another factor suggesting the statement is geared towards forecast or contigency rather than actual fact.
It is not entirely clear either that the 'inventory computers' could corroborate a computer identified in that inventory being the disruptive source, as opposed to merely being a compromised device, thus repetition of events could continue to be propagated against that same device. A reason for that is that an inventory of computers may not take account for 'BYOD devices' (http://www.trewmte.blogspot.co.uk/2013/01/smartphone-byod.html), which, quite horrifically, are being proposed for business. Laundering such ideas as relevant, cost-effective (sorry, a cheap way for a company to provide IT/comms without cost to the company) and a must-have, apparently, on employee-centric wish-lists crudely ignores common-sense security policies, practices and procedures. Moreover, BYOD is more likely to increase the chances of ICT (information communications technology) disruption/compromise, which is crudely labelled 'cyber attack/crime'. Cyber attack/crime is itself is being used in an attempt to downgrade the important differences and characteristics between science technologies by applying to them a technology-neutral title, 'cyber' (c.f. Andrea C Simmons comments about cyber wrapping http://www.bcs.org/content/conBlogPost/1861 )
Promulgating the noun/adjective/verb 'cyber-', lauded as the title of the next big threat to the World will in itself generate changes down the line eventually forcing Governments to create localised technology standards to deliberately undermine and reduce the chances of global cyber attacks/crimes occuring based upon common code used by many countries. This perhaps can be illustrated in terms of the enormous economic and fiscal effort that has gone into maintaining the 'Euro' contrasted with the 'British Pound'. Were it the case the Euro actually went down the localised British Pound would still be there. Local individual currency proving a far sounder bet than global common currency. Or looked at from a different angle, if a local individual currency did go down it wouldn't drag down other currencies partners. This tends to reinforce the positive that 'local standards' offer far better security and protection but minimises disruption, whereas global standards create the greater threat of 'one hit, hits all'.