diff --git a/scrapy/trunk/scrapy/contrib/downloadermiddleware/retry.py b/scrapy/trunk/scrapy/contrib/downloadermiddleware/retry.py
index e7ed17ea5..6af0c3f3b 100644
--- a/scrapy/trunk/scrapy/contrib/downloadermiddleware/retry.py
+++ b/scrapy/trunk/scrapy/contrib/downloadermiddleware/retry.py
@@ -29,6 +29,7 @@ from twisted.internet.defer import TimeoutError as UserTimeoutError
 
 from scrapy import log
 from scrapy.core.exceptions import HttpException
+from scrapy.stats import stats
 from scrapy.conf import settings
 
 class RetryMiddleware(object):
@@ -45,6 +46,16 @@ class RetryMiddleware(object):
         if isinstance(exception, self.EXCEPTIONS_TO_RETRY) or (isinstance(exception, HttpException) and (int(exception.status) in self.retry_http_codes)):
             fp = request.fingerprint()
             self.failed_count[fp] = self.failed_count.get(fp, 0) + 1
+            if isinstance(exception, self.EXCEPTIONS_TO_RETRY):
+                exception_name = exception.__name__
+                if exception_name == 'TimeoutError':
+                    if 'defer' in str(exception):
+                        exception_name = 'UserTimeoutError'
+                    else:
+                        exception_name = 'ServerTimeoutError'
+                stats.incpath('%s/conn_error_count/%s' % (spider.domain_name, exception_name))
+            else:
+                stats.incpath('%s/http_error_count/%d' % (spider.domain_name, int(exception.status)))
 
             if self.failed_count[fp] <= self.retry_times:
                 log.msg("Retrying %s (failed %d times): %s" % (request, self.failed_count[fp], exception), domain=spider.domain_name, level=log.DEBUG)